Latest / Wired To Hunt Podcast / Ep. 1057: AI and the Future of Trail Cameras with Darrin Durham

Transcript

Ep. 1057: AI and the Future of Trail Cameras with Darrin Durham · Wired To Hunt Podcast

• 0:01
  Speaker 1: Welcome to the Wired to Hunt podcast, your guide to
• 0:04
  Speaker 1: the White Tail Woods presented by first Light, creating proven
• 0:08
  Speaker 1: versatile hunting apparel for the stand, saddle or blind. First
• 0:13
  Speaker 1: Light Go Farther, Stay Longer, and now your host, Mark Kenyon.
• 0:19
  Speaker 2: Welcome to the Wired to Hunt podcast. This week on
• 0:22
  Speaker 2: the show, I'm joined by Darren Durham, an avid deer
• 0:25
  Speaker 2: hunter and a program manager at Moultrie Mobile, to discuss
• 0:30
  Speaker 2: the fast evolving integration of AI into trail cameras and
• 0:34
  Speaker 2: how we deer hunters can best take advantage of this
• 0:37
  Speaker 2: new technology. All right, welcome back to the Wired to
• 0:44
  Speaker 2: Hunt podcast, brought to you by Moultrie. Today in the show,
• 0:48
  Speaker 2: we are talking trailcams and specifically AI, artificial.
• 0:53
  Speaker 3: Intelligence, and how.
• 0:55
  Speaker 2: This crazy new technology that seems to be all around
• 0:59
  Speaker 2: us and the rest of our lives is now being
• 1:02
  Speaker 2: integrated into deer hunting and trail cameras specifically. You know,
• 1:07
  Speaker 2: this is something that I have conflicting thoughts on when
• 1:10
  Speaker 2: it comes to many of the AI questions of our day, right,
• 1:14
  Speaker 2: How it's maybe taking jobs, How it's how data centers
• 1:18
  Speaker 2: are impacting neighborhoods and wild places. How this technology is
• 1:22
  Speaker 2: impacting all sorts of different things. I've got real concerns,
• 1:27
  Speaker 2: but at the same time, it is a reality of
• 1:29
  Speaker 2: our world and it is now part of everything right
• 1:32
  Speaker 2: AI is part of how we send emails, as part
• 1:34
  Speaker 2: of how we get our news. It's part of how
• 1:38
  Speaker 2: so many different things are done across all parts of life,
• 1:42
  Speaker 2: banking and business and writing and reading and music and
• 1:47
  Speaker 2: your algorithms that develop your Spotify playlist, you name it,
• 1:51
  Speaker 2: it's AI. So what I wanted to do and what
• 1:54
  Speaker 2: I've been thinking about lately is just, you know, trying
• 1:56
  Speaker 2: to understand where this fits in with our hunting world
• 2:00
  Speaker 2: right now. And I think one of the main places
• 2:01
  Speaker 2: that is is with trailcams. So I was asking around
• 2:05
  Speaker 2: trying to find who might the experts be on this,
• 2:07
  Speaker 2: who could speak to this from a position of authority
• 2:10
  Speaker 2: and experience, And who I got to was Darren Durham.
• 2:13
  Speaker 2: He is a program manager at Moultrue Mobile and his job,
• 2:17
  Speaker 2: a big part of what he does is explores how
• 2:20
  Speaker 2: to integrate AI into Moultree's cameras and their application, their
• 2:25
  Speaker 2: app in which they help you or allow you to
• 2:29
  Speaker 2: analyze and review and make predictions and learn all sorts
• 2:32
  Speaker 2: of stuff based on those trailcam pictures and both sides
• 2:36
  Speaker 2: of that, the cameras and the app. There is really
• 2:39
  Speaker 2: deep AI integration, it turns out, and this is the
• 2:42
  Speaker 2: case across many other camera companies too.
• 2:45
  Speaker 3: This isn't just Moultrie.
• 2:46
  Speaker 2: So what I want to get Darren on here to
• 2:49
  Speaker 2: discuss is where is this what does this look like?
• 2:53
  Speaker 2: And not just for Moultrue, but for many different camera companies.
• 2:56
  Speaker 2: How can we as deer hunters best use this stuff?
• 2:59
  Speaker 2: Where does it makes sense to use it? How can
• 3:01
  Speaker 2: it help us as hunters? How can it help us
• 3:03
  Speaker 2: take better photos or better understand our photos, or save
• 3:06
  Speaker 2: battery life on our cameras, anything like that. These are
• 3:09
  Speaker 2: the types of questions I eventually wanted to get out
• 3:12
  Speaker 2: here was really the application side of this. What do
• 3:14
  Speaker 2: we need to know, what don't we need to know?
• 3:16
  Speaker 2: And how do we do this stuff? That is what
• 3:19
  Speaker 2: I cover today with Darren. If you use trail cameras
• 3:22
  Speaker 2: of any kind, there should be information here that can
• 3:24
  Speaker 2: help you help you understand how this technology is changing,
• 3:29
  Speaker 2: what can be done right now, and then also interestingly,
• 3:31
  Speaker 2: what might be available here in the coming years. That's
• 3:34
  Speaker 2: another final thing that I pressed Darren on is what's
• 3:37
  Speaker 2: the future going to hold?
• 3:38
  Speaker 3: So hopefully you guys will find this one interesting.
• 3:41
  Speaker 2: Again, I think that there's plenty here to explore, regardless
• 3:45
  Speaker 2: of the camera company you choose to use. Darren, of
• 3:48
  Speaker 2: course is from Moultrie. That's something I've got experiences with
• 3:51
  Speaker 2: as well. But I don't care what team you play
• 3:55
  Speaker 2: for when it comes to the cameras. I think you're
• 3:57
  Speaker 2: going to find this one interesting and I hope you do.
• 3:59
  Speaker 2: So thanks for joining us here.
• 4:06
  Speaker 3: All right with me on the line now, is Darren Durham.
• 4:09
  Speaker 3: Welcome the show.
• 4:09
  Speaker 4: Darren. Thanks happy to be here.
• 4:12
  Speaker 2: I really appreciate you making the time, especially here on
• 4:15
  Speaker 2: a Friday afternoon in the summer. I know you've definitely
• 4:18
  Speaker 2: got better things going on than this, so thanks for
• 4:21
  Speaker 2: making the televion.
• 4:21
  Speaker 4: Then it's like ninety to one hundred degrees outside, so.
• 4:24
  Speaker 2: Fair enough, this might be actually a good excuse than
• 4:27
  Speaker 2: to stay in the air conditioning.
• 4:28
  Speaker 4: Right.
• 4:30
  Speaker 2: So I wanted to talk to you, Darren, because I
• 4:34
  Speaker 2: have a set of questions and curiosities related to something
• 4:39
  Speaker 2: that I just keep.
• 4:40
  Speaker 3: Hearing about, keep reading about, keep.
• 4:43
  Speaker 2: Wondering about, over and over and over, and I don't
• 4:46
  Speaker 2: think I'm the only one on this topic, because AI,
• 4:49
  Speaker 2: artificial intelligence, it's you know, for the last four years
• 4:53
  Speaker 2: or so, it has been absolutely you know, a fury
• 4:57
  Speaker 2: of news and changes and he in hysteria, you know,
• 5:01
  Speaker 2: across all aspects of life, right, Yeah, And I feel
• 5:06
  Speaker 2: like this is one of those things that some people
• 5:09
  Speaker 2: are hyping it up is the best thing since sliced bread,
• 5:12
  Speaker 2: and then other people are talking about how this is awful,
• 5:15
  Speaker 2: and there's all these concerns and worries about it, and interestingly,
• 5:19
  Speaker 2: there's this set of ways that it's starting to infiltrate
• 5:22
  Speaker 2: into the hunting world that's gotten me curious about, Okay,
• 5:25
  Speaker 2: you know all of those other things aside, how does
• 5:29
  Speaker 2: this stuff impact what we do as hunters?
• 5:31
  Speaker 3: And I've seen it.
• 5:32
  Speaker 2: Popping up more and more in relation to traial cameras,
• 5:35
  Speaker 2: and so that's why I wanted to talk to you
• 5:36
  Speaker 2: as someone who's actually been involved in that side of
• 5:40
  Speaker 2: AI use and application. And so what I wanted to
• 5:43
  Speaker 2: start with was kind of high level, Darren, when you
• 5:47
  Speaker 2: think about all of this, when you think about you know,
• 5:50
  Speaker 2: all of this hype, all of this hysteria, good and bad,
• 5:54
  Speaker 2: and you look at it with what you know and
• 5:56
  Speaker 2: the way you've been involved in it.
• 5:57
  Speaker 3: How are you thinking about it?
• 5:59
  Speaker 2: You do you look at AI, you know, in the
• 6:01
  Speaker 2: context of hunting and trail camera applications and kind of
• 6:04
  Speaker 2: this world we're in.
• 6:06
  Speaker 3: Would you say that AI seems to be overrated and
• 6:09
  Speaker 3: overhyped right now?
• 6:10
  Speaker 2: Or do you think that it's it's it is and
• 6:13
  Speaker 2: or it's going to be everything that's hyped up to be,
• 6:16
  Speaker 2: and that maybe we're just scratching the surface.
• 6:19
  Speaker 5: Well that's I mean, that's obviously a tough question. But
• 6:22
  Speaker 5: I do think that there's some parts of it we're
• 6:24
  Speaker 5: touching the surface of. I think there are other parts
• 6:26
  Speaker 5: of it that, you know, we we've pretty much gotten
• 6:29
  Speaker 5: what we could get out of it. I mean, one
• 6:31
  Speaker 5: way to you know, frame up AI on a camera
• 6:34
  Speaker 5: real quick is that you know, there's there's two different
• 6:38
  Speaker 5: things with AI. There's the device or the the app
• 6:42
  Speaker 5: itself actually classifying images or tagging images from what the
• 6:51
  Speaker 5: artificial intelligence sees.
• 6:53
  Speaker 4: It's object detection of some sort.
• 6:55
  Speaker 5: It's sees something in the picture, it tags it with
• 6:58
  Speaker 5: what it's all. And then there's obviously the aggregated data
• 7:01
  Speaker 5: you know that they does get lumped together as AI,
• 7:04
  Speaker 5: you know, and like pretty much.
• 7:06
  Speaker 4: Every app these days has some sort of way for
• 7:12
  Speaker 4: you to filter pictures.
• 7:13
  Speaker 5: I only want to see Bucks, or I only want
• 7:15
  Speaker 5: to see Deer, or I only want to see Bucks
• 7:18
  Speaker 5: from on this day, from twelve to three. You know,
• 7:23
  Speaker 5: like like all of that is is now. It's a
• 7:26
  Speaker 5: reality for us. So back to your to answer your question,
• 7:31
  Speaker 5: you know, knowing those two things, I think we're just
• 7:33
  Speaker 5: scratching the surface of how the data gets aggregated and
• 7:37
  Speaker 5: delivered to us. You know, like what else can we
• 7:39
  Speaker 5: scrape out of a huge data set that you know,
• 7:43
  Speaker 5: we're as a as a camera company compiling for you
• 7:46
  Speaker 5: as a customer, you know. And then the technology itself,
• 7:50
  Speaker 5: which is like the the root of how it works,
• 7:53
  Speaker 5: which is the one way to think about it, is
• 7:56
  Speaker 5: like the it's kind of like an an ARC, you
• 7:59
  Speaker 5: know that it's detect, classify, act, and predict, and each
• 8:03
  Speaker 5: one of those is a phase that AI will help
• 8:05
  Speaker 5: you do, whether that AI is on the device or
• 8:08
  Speaker 5: that AI is encapsulated somewhere in in the in a
• 8:12
  Speaker 5: server in a cloud server somewhere that you know, delivers
• 8:15
  Speaker 5: it to you via via an app somewhere so so anyway,
• 8:20
  Speaker 5: just bring it back around. I think it's I think
• 8:22
  Speaker 5: we're just scratching the surface of of of the data
• 8:25
  Speaker 5: aggregation and data delivery to the customer. I think the
• 8:30
  Speaker 5: tech world kind of drives some of the other, you know,
• 8:34
  Speaker 5: what the hardware self is capable of. But I think
• 8:37
  Speaker 5: we're just scratching the surface of that too. You out actuality,
• 8:41
  Speaker 5: how do you guys? Yeah, I know you mentioned a
• 8:44
  Speaker 5: couple of different phases there. There seems to be the
• 8:47
  Speaker 5: you know, detection, and then there's a analysis maybe ye
• 8:52
  Speaker 5: when it comes to like teaching an AI or somehow
• 8:55
  Speaker 5: getting this tool to be able to identify an animal
• 9:00
  Speaker 5: or identify some detail, like how does how do you
• 9:03
  Speaker 5: guys teach that stuff? Like?
• 9:04
  Speaker 2: I understand how when I think of like chat, GPT
• 9:07
  Speaker 2: or something. Right, they teach these large language models by
• 9:11
  Speaker 2: feeding them millions and millions of data points of all
• 9:14
  Speaker 2: of the texts that we've ever created on the Internet,
• 9:17
  Speaker 2: and then you know, these models then use that to
• 9:20
  Speaker 2: understand and predict what's you know, the next thing should be?
• 9:23
  Speaker 3: Right? How does that work? When it comes to like
• 9:26
  Speaker 3: photos or.
• 9:27
  Speaker 2: You know, an animal, well, I have no idea how
• 9:30
  Speaker 2: that might look is exact same.
• 9:32
  Speaker 5: Well exact same, not exact same, but the method or
• 9:37
  Speaker 5: the methodology or take you know, technology, whatever you want
• 9:40
  Speaker 5: to call it, Like, the direction behind it is is
• 9:43
  Speaker 5: that you know, we we have thousands of images taken
• 9:47
  Speaker 5: by a device, by a camera. I guess I shouldn't
• 9:49
  Speaker 5: say advice, but taken by a camera could be a device,
• 9:51
  Speaker 5: but taken by a camera. Those thousands of images get
• 9:54
  Speaker 5: put in and annotated, which means that somebody physically sits
• 9:58
  Speaker 5: there and says yes, but yes deer, yes, kyo, yes.
• 10:03
  Speaker 4: Duck whatever.
• 10:04
  Speaker 5: Whatever it is that that AI engine is looking or
• 10:07
  Speaker 5: that object detection model is looking for, there's somebody behind
• 10:11
  Speaker 5: there at some point that has trained thousands of images
• 10:15
  Speaker 5: in order to create that network of ability essentially for
• 10:20
  Speaker 5: you know, the AI engine to say, yep, I am
• 10:23
  Speaker 5: eighty percent confident that is a deer based off of
• 10:26
  Speaker 5: the training that I've had.
• 10:28
  Speaker 4: So you know, like.
• 10:31
  Speaker 5: We're we're pretty accurate, I would say, you know, like
• 10:34
  Speaker 5: we're within We're ninety plus percent accurate on a lot
• 10:37
  Speaker 5: of images. You know, But it's all about the training.
• 10:42
  Speaker 5: You know, if if that model's never seen a deer
• 10:45
  Speaker 5: with a feeder in the background. It's it's a higher
• 10:49
  Speaker 5: likelihood to get that wrong than you know, a model
• 10:53
  Speaker 5: that has seen ten thousand deer with a feeder in
• 10:57
  Speaker 5: the background, or with a particular are looking limb in
• 11:01
  Speaker 5: the background that makes it look like it has horns
• 11:03
  Speaker 5: or something like that. So I mean, it's it's It
• 11:06
  Speaker 5: really just comes down to, like the camera.
• 11:11
  Speaker 4: The same technology that.
• 11:13
  Speaker 5: Has been used for years to tell the camera to
• 11:16
  Speaker 5: take a picture as being used, but what's happening after
• 11:19
  Speaker 5: that camera snaps that image. It's completely different now, so
• 11:23
  Speaker 5: that that camera essentially goes and looks across that entire
• 11:27
  Speaker 5: image and says, do I see anything that looks like something?
• 11:31
  Speaker 5: And then it then the model says, oh, yeah, I
• 11:34
  Speaker 5: see three things in there that could be something.
• 11:36
  Speaker 4: Let's take, for example, like.
• 11:37
  Speaker 5: An image that has a feeder in it, a deer
• 11:40
  Speaker 5: in it, and I don't know, say a tall stump
• 11:43
  Speaker 5: in it.
• 11:44
  Speaker 4: That that object that object detection model.
• 11:46
  Speaker 5: Might or the AI whatever whatever you want to call it.
• 11:50
  Speaker 5: The technology sees that there are three different things. It
• 11:53
  Speaker 5: sees a stump, it sees a deer, and it sees
• 11:56
  Speaker 5: a feeder. It knows nothing of any of the.
• 11:58
  Speaker 4: Three at first pass.
• 12:01
  Speaker 5: So it draws a box around all of them, and
• 12:03
  Speaker 5: then it says, okay, I've saw I saw three three things.
• 12:05
  Speaker 5: I need to go look at those three things and
• 12:07
  Speaker 5: see if they're in my training set.
• 12:09
  Speaker 4: And then it goes it says.
• 12:10
  Speaker 5: Well, I've never Darren didn't tell me I needed to
• 12:12
  Speaker 5: look for a stump, you know, Darren didn't tell me
• 12:15
  Speaker 5: I needed to look for a feeder up.
• 12:17
  Speaker 4: But there's a deer. Darren told me I needed to
• 12:18
  Speaker 4: look for a deer.
• 12:19
  Speaker 5: And then it kind of runs through its bank of
• 12:22
  Speaker 5: you know, if deer, then what is it a buck?
• 12:26
  Speaker 4: Does it have horns? Does it not have horns?
• 12:28
  Speaker 5: You know? And and I mean, you know, we're there
• 12:32
  Speaker 5: were some things are being done now to like for
• 12:37
  Speaker 5: like future future, to move it, move it further than
• 12:40
  Speaker 5: the future, which are like is it a mature buck?
• 12:43
  Speaker 5: You know?
• 12:43
  Speaker 4: Is it a small buck? Is it a fall?
• 12:46
  Speaker 5: You know?
• 12:46
  Speaker 4: Whatever?
• 12:47
  Speaker 3: You know?
• 12:47
  Speaker 4: But but all in all, that's how it works.
• 12:49
  Speaker 5: It Essentially, the device takes a picture like it always has,
• 12:52
  Speaker 5: and then it runs through this big post program that
• 12:55
  Speaker 5: says do I see anything in the picture?
• 12:56
  Speaker 4: And what are those things I see in the picture?
• 12:59
  Speaker 2: Is there you mentioned that you're going and training, you know,
• 13:04
  Speaker 2: based on somebody actually annotating photos. Is there any kind
• 13:08
  Speaker 2: of course correction after analysis too? Like, for example, sure
• 13:13
  Speaker 2: we would go in and say, okay, the program said
• 13:16
  Speaker 2: that this was a deer, a big we'll say, a
• 13:19
  Speaker 2: mature buck like you mentioned, but actually the human looks
• 13:22
  Speaker 2: at it and it's it's a young.
• 13:23
  Speaker 3: Buck with a big tree branch behind it. That makes
• 13:25
  Speaker 3: it look bigger. Right. Is there any of that kind
• 13:28
  Speaker 3: of like teaching post happening as well.
• 13:31
  Speaker 4: A lot of a lot of post teaching.
• 13:32
  Speaker 5: Actually a couple of ways that that might that that
• 13:36
  Speaker 5: might we might do that. Here's you know, one of us,
• 13:40
  Speaker 5: you know, we're we're heavy users, you know, like any
• 13:42
  Speaker 5: of us, multiple cameras, multiple accounts, We're constantly looking at them,
• 13:47
  Speaker 5: and you know, if we see something, it's like why
• 13:49
  Speaker 5: does this thing keep returning dose when I'm filtering, you know.
• 13:52
  Speaker 5: And that's probably another thing I should add before we
• 13:54
  Speaker 5: go into that question, is what the AI really allows
• 13:57
  Speaker 5: the customer to use or the customer to do, is
• 14:00
  Speaker 5: filter easily in the in the gallery, you know, like
• 14:03
  Speaker 5: I don't have to look at the one hundred raccoon
• 14:06
  Speaker 5: pictures that I got or the you know, whatever I don't,
• 14:09
  Speaker 5: I essentially see what I want to see by just
• 14:12
  Speaker 5: moving a filter.
• 14:13
  Speaker 4: So when you're looking through your gallery of images and you.
• 14:16
  Speaker 5: Filter it for bucks, let's say bucks as well as
• 14:19
  Speaker 5: you used the buck, you know, and I see Doze
• 14:22
  Speaker 5: predominantly Doze. You know, I might go back to our
• 14:25
  Speaker 5: engineering team and say, hey, man, like the last two days,
• 14:29
  Speaker 5: like for some reason, the model's pretty bad off, you know,
• 14:32
  Speaker 5: and then they'll look at it and they'll say, oh,
• 14:34
  Speaker 5: it's because you got a feeder in the background.
• 14:36
  Speaker 4: You know, we haven't trained enough images.
• 14:38
  Speaker 5: We'll out of our ten thousand image set that we
• 14:42
  Speaker 5: ran through, only one hundred of them had a feeder
• 14:45
  Speaker 5: in it. We now we got to go, well, we
• 14:47
  Speaker 5: don't have to, but we would go build a set
• 14:50
  Speaker 5: of images that has feeder.
• 14:52
  Speaker 4: Ten thousand images that have feeders in it. You know.
• 14:55
  Speaker 5: That's how the model improves is by you know, you
• 14:57
  Speaker 5: identify it kind of a chink in the armor, and
• 15:00
  Speaker 5: then you just expose it. You know, it's like, Okay,
• 15:02
  Speaker 5: there's something going on here, maybe not sure what that is,
• 15:06
  Speaker 5: but we're we're gonna dig into it, and then we're
• 15:08
  Speaker 5: gonna spend another Another version of the model.
• 15:11
  Speaker 4: We're gonna run it.
• 15:12
  Speaker 5: We're gonna look at it and see is it increasing
• 15:14
  Speaker 5: or decreasing. We monitor what they call IF one scores
• 15:18
  Speaker 5: all the time, which is essentially the health of the
• 15:21
  Speaker 5: object detection model, or not really the health, but more
• 15:24
  Speaker 5: of the accuracy of the model.
• 15:26
  Speaker 4: We monitor those scores a lot.
• 15:27
  Speaker 5: When we start to see those scores kind of come
• 15:30
  Speaker 5: down based off of our benchmark, you know, we might
• 15:33
  Speaker 5: we might go back and retrain and evaluate where our
• 15:36
  Speaker 5: where our weaknesses are to improve the customer experience.
• 15:40
  Speaker 2: So on this, on this object recognition side of things,
• 15:45
  Speaker 2: is this becoming commonplace across trail camera brands? Is this
• 15:50
  Speaker 2: unique to you guys still or is this where everything's headed?
• 15:53
  Speaker 4: Yeah, that's that's a that's a good question.
• 15:56
  Speaker 5: So there are different levels of object detection or recognition,
• 16:00
  Speaker 5: whatever you want to call it. There are different levels
• 16:02
  Speaker 5: of that. To our previous point early, you know, like
• 16:06
  Speaker 5: we are making advancements in that realm, you know, and
• 16:10
  Speaker 5: and those advancements, you know, like generally speaking, I guess,
• 16:14
  Speaker 5: let me simplify a little bit, slow down.
• 16:16
  Speaker 4: I guess for myself to make sure everybody kind of
• 16:19
  Speaker 4: gets it.
• 16:19
  Speaker 5: So what what everybody gets generally for free. Is the
• 16:25
  Speaker 5: ability to filter by object. You know, let me filter
• 16:29
  Speaker 5: all my images to deer, let me filter all my
• 16:31
  Speaker 5: images to Turkey, whatever. The part that they generally don't
• 16:36
  Speaker 5: get for free, it's generally.
• 16:38
  Speaker 4: Behind a paywall.
• 16:39
  Speaker 5: Is that kind of prediction, you know, like in prediction
• 16:44
  Speaker 5: may be kind of a.
• 16:47
  Speaker 4: Way out there word.
• 16:48
  Speaker 5: I mean, yes, we do predict some things based off
• 16:51
  Speaker 5: of you know, certain data points and things like that
• 16:53
  Speaker 5: that we're gathering. But that part piece of things, and
• 16:56
  Speaker 5: we can talk about that in the future whenever you're ready.
• 16:59
  Speaker 5: But that part of it is generally what costs, you know,
• 17:03
  Speaker 5: what what the customer has to you know, pay extra
• 17:06
  Speaker 5: for or you know, upgrade something in the in the plan.
• 17:10
  Speaker 5: Sometimes you know, if you go to unlimited, you might
• 17:12
  Speaker 5: get it for free or something, but the data plans anyway.
• 17:15
  Speaker 4: But yes, generally speaking, the object detection piece.
• 17:19
  Speaker 5: Is free on the app, and I guess it's but
• 17:22
  Speaker 5: it's it's not every device does not have it, so
• 17:26
  Speaker 5: I guess that's a good separation point, right, So devices
• 17:30
  Speaker 5: also run those object detection models in the field. So
• 17:35
  Speaker 5: up until you know, for Moultrie, it was it was
• 17:38
  Speaker 5: the edge pro which was like a three year three.
• 17:41
  Speaker 4: The camera was like three years ago. So that was
• 17:43
  Speaker 4: our first device to.
• 17:44
  Speaker 5: Carry onboard AI or onboard object detection. So what that
• 17:51
  Speaker 5: means is that device was doing all of that compute
• 17:53
  Speaker 5: that we were just talking about.
• 17:55
  Speaker 4: You know, is there something in the picture? What is
• 17:57
  Speaker 4: that something? It was?
• 17:59
  Speaker 5: That device was those All of our devices, starting with
• 18:02
  Speaker 5: the Edge Pro, are now.
• 18:04
  Speaker 4: Capable of doing that.
• 18:06
  Speaker 5: And what that does for the customer is it essentially
• 18:08
  Speaker 5: saves them battery, you know, some some data cost. The biggest,
• 18:12
  Speaker 5: biggest piece of that is the battery savings. You know,
• 18:14
  Speaker 5: your device now is only only uploading what you want
• 18:19
  Speaker 5: to see, meaning I only want to see deer. I
• 18:22
  Speaker 5: don't care about turkeys or anything else right now. I
• 18:25
  Speaker 5: just want to see deer. So my device is all
• 18:29
  Speaker 5: they send me are deer pictures. You know, we've taken
• 18:31
  Speaker 5: it to the next level and also said like all
• 18:34
  Speaker 5: I want to see is bucks. I don't even care
• 18:36
  Speaker 5: about those, you know, like I maybe I don't have
• 18:38
  Speaker 5: a dough tag, maybe there's no dough seasons out in
• 18:42
  Speaker 5: that particular state or whatever, you know, but like now
• 18:44
  Speaker 5: you can you can really drill down per device and
• 18:49
  Speaker 5: tell that device exactly what you want it to do.
• 18:51
  Speaker 5: I mean, we we take it as far as you know,
• 18:53
  Speaker 5: like a object based settings too. Now you know, multially
• 18:56
  Speaker 5: allows you to say, you know, like hey, for for
• 19:00
  Speaker 5: every picture other than a buck deer, send me a
• 19:05
  Speaker 5: steal and upload them once a day. But for bucks,
• 19:08
  Speaker 5: I want you to take a fifteen second video and upload.
• 19:11
  Speaker 4: It immediately, like we can.
• 19:14
  Speaker 5: We are it is available to the customer today on
• 19:17
  Speaker 5: the multi side where we can actually you know, the
• 19:20
  Speaker 5: customer gets what they want when they want it, essentially
• 19:23
  Speaker 5: in holling.
• 19:24
  Speaker 3: Yeah, so that's pretty coraod.
• 19:25
  Speaker 5: Eaving tons of battery, you know, like it'll, it'll, it
• 19:28
  Speaker 5: keeps your devices in the field for longer.
• 19:40
  Speaker 2: Could you ever get to the point and I maybe
• 19:43
  Speaker 2: there is a way to do this already and I
• 19:44
  Speaker 2: just don't realize it, But would there ever be a
• 19:47
  Speaker 2: way to do some of that specific kind of setting
• 19:49
  Speaker 2: stuff to the individual buck level, Like are we to
• 19:53
  Speaker 2: the point where we can train the algorithm to target
• 19:57
  Speaker 2: a specific deer, like, Hey, this deer that I tagged
• 20:00
  Speaker 2: as the white eight, could you automatically send me photos
• 20:05
  Speaker 2: of that deer or give me an HD funnel automatically
• 20:08
  Speaker 2: of that deer or any of these other things like
• 20:10
  Speaker 2: are we to that point?
• 20:12
  Speaker 5: We are not quite there yet, or maybe I should
• 20:16
  Speaker 5: say we're not there yet, But that is the.
• 20:22
  Speaker 4: That's the code, Gras. You know what I mean when
• 20:24
  Speaker 4: I can say, show.
• 20:26
  Speaker 5: Me the hammer, you know, year over year, yeah, you know,
• 20:31
  Speaker 5: and like it's just putting it together, you know, and
• 20:33
  Speaker 5: then you know, I mean as you throw the prediction
• 20:36
  Speaker 5: piece on top of it, you know, and like you know,
• 20:39
  Speaker 5: there there could be a day where you know, AI
• 20:42
  Speaker 5: or objecttation, whatever you want to call it, it sees
• 20:44
  Speaker 5: a deer that it knows from you know, from all
• 20:48
  Speaker 5: of that training that we've given it, that that deer
• 20:50
  Speaker 5: is gonna be a stud in three years.
• 20:53
  Speaker 4: I'm gonna watch him.
• 20:54
  Speaker 5: I'm gonna tell you everything I know about him, and
• 20:57
  Speaker 5: then I'm just gonna deliver, you know, like I'm gonna Essentially,
• 21:02
  Speaker 5: it's doing the things that we're doing.
• 21:03
  Speaker 4: Mark.
• 21:04
  Speaker 5: You know, we we spend a ton of time combing
• 21:07
  Speaker 5: through pictures, combing through weather, combing through all kinds of.
• 21:12
  Speaker 3: So many spreadsheets of stuff.
• 21:16
  Speaker 5: You know, like that's where the benefit comes in, you know,
• 21:18
  Speaker 5: that's that's again, that's that's where I think it'll be
• 21:22
  Speaker 5: in the next you know, like three to.
• 21:25
  Speaker 4: Five years, let's say.
• 21:26
  Speaker 5: I mean, obviously I threw that number out there, and
• 21:28
  Speaker 5: I don't know if it'll be three to five years,
• 21:29
  Speaker 5: it might be two months. I don't know, but I
• 21:32
  Speaker 5: definitely think that's the direction it's going is to give you.
• 21:35
  Speaker 4: All of that.
• 21:36
  Speaker 2: Okay, So so staying with the object detection side of things,
• 21:40
  Speaker 2: just to continue to kind of drill down on that side.
• 21:43
  Speaker 2: You mentioned that there's the on board side of it,
• 21:46
  Speaker 2: in which the actual device itself can analyze what's in
• 21:50
  Speaker 2: front of it right now. And is that happening like
• 21:53
  Speaker 2: in the moment or does it does it trigger as
• 21:57
  Speaker 2: I guess I'm saying, like, let's said, deer walks in
• 21:59
  Speaker 2: front of the camp, Yeah, it triggers. Is it then
• 22:02
  Speaker 2: doing this object analysis after that photo has been triggered?
• 22:06
  Speaker 2: It is now it's on board, and then the camera
• 22:10
  Speaker 2: analyze it and says upload or don't upload. I mean,
• 22:13
  Speaker 2: it couldn't possibly be doing that object detection process as
• 22:16
  Speaker 2: that deer is passing through and making a decision.
• 22:20
  Speaker 4: It's amazing.
• 22:21
  Speaker 5: It's it's doing it in milliseconds. It is actually not
• 22:24
  Speaker 5: even doing it. It's actually not even doing it on
• 22:29
  Speaker 5: like a I guess, a full resolution image. You might say,
• 22:32
  Speaker 5: like it's doing it in a very very early in
• 22:36
  Speaker 5: the process, like yes, the pr says, camera, wake up,
• 22:41
  Speaker 5: there's something out there. But the camera doesn't fire up
• 22:44
  Speaker 5: all of its systems until it keeps a very like
• 22:48
  Speaker 5: a skeleton crew running where it could see just enough
• 22:51
  Speaker 5: of that image to know, Yep, that's a deer, and
• 22:54
  Speaker 5: yes it's a buck because we've trained it to.
• 22:58
  Speaker 4: So it's using a very low resolution.
• 23:00
  Speaker 5: Image to say, yes, that's a book, take a picture
• 23:03
  Speaker 5: of it, or yes that's a book, take a video
• 23:05
  Speaker 5: of it.
• 23:05
  Speaker 4: So that's where you're getting the battery savings. You're not.
• 23:08
  Speaker 5: All of your battery savings doesn't come from just the
• 23:11
  Speaker 5: device uploading or not uploading the It's really the entire system,
• 23:16
  Speaker 5: you know, like I don't. The device doesn't turn on
• 23:19
  Speaker 5: certain subsystems if it deems it a null capture. It
• 23:24
  Speaker 5: just leaves everything in idle state and waits for another one.
• 23:29
  Speaker 4: Wow.
• 23:30
  Speaker 3: Okay, so.
• 23:34
  Speaker 2: How does someone how does the user make sure that
• 23:37
  Speaker 2: they are fully enjoying that benefit that the battery saving
• 23:40
  Speaker 2: side of this. It's kind of the way I'm thinking
• 23:43
  Speaker 2: through this is like I want to understand each slice
• 23:45
  Speaker 2: of the pie here, like the application, and then I
• 23:48
  Speaker 2: want to understand, Okay, how do we make sure we're
• 23:49
  Speaker 2: doing this to take advantage of that benefit? So we
• 23:53
  Speaker 2: just we just covered AI will actually detect onboard immediately
• 23:58
  Speaker 2: and any either fire up or not. So what will
• 23:59
  Speaker 2: I need to do to make sure I am you know,
• 24:02
  Speaker 2: have I'm utilizing that?
• 24:05
  Speaker 5: For Multrie, you would go into your capture settings, and
• 24:08
  Speaker 5: within capture settings, there's a general settings bucket and then
• 24:12
  Speaker 5: there's a there's a smart capture bucket, and you really
• 24:17
  Speaker 5: just essentially you can go into your smart capture bucket
• 24:19
  Speaker 5: and you tell it what you want if you you
• 24:23
  Speaker 5: tell it what you want to see, you know, and
• 24:25
  Speaker 5: a lot of people I think you know to your
• 24:27
  Speaker 5: point previously, you know, it's a little the confidence might
• 24:30
  Speaker 5: wane a little bit. You know, it's we all have it.
• 24:33
  Speaker 5: It's technology. You know, until you've been using it for
• 24:36
  Speaker 5: twenty years, you don't tend to rely on it. But yeah,
• 24:41
  Speaker 5: you just go in and tell it what you want
• 24:43
  Speaker 5: and how you want it.
• 24:44
  Speaker 4: And that I mean, it's it's simple, it's super simple.
• 24:50
  Speaker 2: What is your confidence rating? I know you can't speak
• 24:53
  Speaker 2: to anybody any other brands who are doing their own
• 24:55
  Speaker 2: version of this, but at least from what you have
• 24:57
  Speaker 2: built and what you've worked on with Moultrie, how confident
• 25:00
  Speaker 2: do you feel in those you know on the accuracy there,
• 25:03
  Speaker 2: because because I've I've looked at this and thought to myself,
• 25:07
  Speaker 2: it seems amazing. Yeah, but I would be really disappointed
• 25:12
  Speaker 2: if that big buck I've been chasing all season came
• 25:15
  Speaker 2: through and was way in the back of the frame
• 25:17
  Speaker 2: or in the back corner or something, and the AI
• 25:20
  Speaker 2: did not detect it and so chose not to take
• 25:22
  Speaker 2: that picture, when if it had been an old camera
• 25:26
  Speaker 2: that I would have gone through every one of these
• 25:27
  Speaker 2: and looked at I would have seen him in the
• 25:28
  Speaker 2: corner and this would have been a great piece of information.
• 25:32
  Speaker 3: How do you consider that? How do you think about
• 25:35
  Speaker 3: that yourself?
• 25:35
  Speaker 5: That's a that's actually a great example. I actually noted
• 25:38
  Speaker 5: that as a as a conversation topic too. So that's
• 25:42
  Speaker 5: where things get. So to answer your first question, ninety
• 25:45
  Speaker 5: five percent certain that I'm going to the multie my
• 25:49
  Speaker 5: Moultarie cameras, whether as an owner or a builder manager,
• 25:54
  Speaker 5: whatever you know, like Moultie cameras are going to be,
• 25:57
  Speaker 5: they're going to be ninety percent accurate at what they
• 26:00
  Speaker 5: can see and what they know. That's where we come
• 26:02
  Speaker 5: into play is to say, have we trained our model
• 26:05
  Speaker 5: good enough with enough negative stuff.
• 26:09
  Speaker 4: So it's ninety ninety five percent.
• 26:11
  Speaker 5: Now that that example that you had where there's you know,
• 26:14
  Speaker 5: like you know, oh, sad, Daddy's back in the you know,
• 26:18
  Speaker 5: the left corner, you know, barely out of reach of
• 26:21
  Speaker 5: the flash. Yeah, you know, like I mean we might
• 26:24
  Speaker 5: catch him. I would say that I'm sixty percent confident.
• 26:28
  Speaker 4: That we would catch that that one video.
• 26:32
  Speaker 5: So really like that's the only place is those really
• 26:37
  Speaker 5: edge cases, you know, It's it's really hard to teach
• 26:40
  Speaker 5: that technology to identify the edge cases. Now. I think
• 26:44
  Speaker 5: in the next couple of years, I think we'll be
• 26:45
  Speaker 5: picking that deer up in the the shadows on the left,
• 26:50
  Speaker 5: the far out left side, you know, cameras, our cameras
• 26:54
  Speaker 5: like we we eliminate one hundred feet now or p
• 26:56
  Speaker 5: i RS capture out to one hundred feet, you know,
• 26:58
  Speaker 5: And that's that's pretty much the market got to do
• 27:00
  Speaker 5: that or you're not a player, you know. And and
• 27:02
  Speaker 5: I think we'll push a I to do better at
• 27:05
  Speaker 5: what our take what our hardware is able to do.
• 27:08
  Speaker 2: Yeah, okay, all right, so we use that smart detect setting,
• 27:14
  Speaker 2: we can choose, you know, what kind of pictures we
• 27:17
  Speaker 2: want to get. Is there any kind of middle ground
• 27:20
  Speaker 2: with that in which I want to utilize some of
• 27:24
  Speaker 2: these AI features to reduce my battery use or something
• 27:28
  Speaker 2: like that, But at the same time, I don't want to.
• 27:31
  Speaker 3: Yeah, I'm worried about what we just talked about. I'm
• 27:33
  Speaker 3: worried about that edge case, you know.
• 27:36
  Speaker 2: Can I think that what I've kind of looked at
• 27:38
  Speaker 2: in the past, is you know, using the object detection
• 27:41
  Speaker 2: to do you know, I think you guys have a
• 27:43
  Speaker 2: feature in which, like you'll auto you know, deliver a
• 27:48
  Speaker 2: video I think on a buck. But I'm still gonna
• 27:50
  Speaker 2: get all my pictures. But I'm just going to get
• 27:52
  Speaker 2: better pictures for bucks or better you know, assets when
• 27:55
  Speaker 2: it's a buck. Is that the extent of the way
• 27:57
  Speaker 2: to kind of modify and find a middle ground or
• 27:59
  Speaker 2: is there any other that you guys are saying, Hey,
• 28:01
  Speaker 2: here's here's a way you can get a better kind
• 28:04
  Speaker 2: of experience for certain things. But you if you want
• 28:07
  Speaker 2: to see it all, if you want to double check everything,
• 28:09
  Speaker 2: you still have that option.
• 28:11
  Speaker 4: Yeah, yeah, those are for MULTU.
• 28:13
  Speaker 5: That's a smart capture uh smart capture object based settings,
• 28:17
  Speaker 5: which essentially that's that's our you know, proprietary name order.
• 28:20
  Speaker 5: That's what you see it as in the app, is
• 28:22
  Speaker 5: smart capture. That's our that's our on device AI and
• 28:25
  Speaker 5: it essentially does exactly what you said, you know. That's
• 28:28
  Speaker 5: what I was speaking on earlier, is that you can
• 28:31
  Speaker 5: tell it.
• 28:32
  Speaker 4: I think we have. I don't don't quote me, I don't.
• 28:35
  Speaker 4: I don't recall how.
• 28:35
  Speaker 5: Many containers that we have and want to say containers,
• 28:38
  Speaker 5: how many options we have in there, but we have
• 28:41
  Speaker 5: definitely more than deer, and you can tell it, you know, like, hey,
• 28:44
  Speaker 5: I want all of these and I want them to
• 28:46
  Speaker 5: upload immediately or in a different fashion.
• 28:49
  Speaker 4: I want them to be videos.
• 28:50
  Speaker 5: I want them to be you know, I don't whatever
• 28:53
  Speaker 5: other you know, image setting you might want, and then
• 28:57
  Speaker 5: everything else. I still want those for what we were
• 29:00
  Speaker 5: just talking about, that big monster that was back in
• 29:03
  Speaker 5: the shadows. You know, you can still upload everything and
• 29:06
  Speaker 5: you can set it to upload at a different cadence
• 29:08
  Speaker 5: that also saves battery, you know, like an immediate upload
• 29:12
  Speaker 5: every image uploading immediately. You can look at it like capture.
• 29:15
  Speaker 5: It's the cost of the capture and the upload for
• 29:18
  Speaker 5: every image, where when you bundle them all together, like
• 29:21
  Speaker 5: all of these kind of secondary and tertiary images that
• 29:25
  Speaker 5: I think.
• 29:26
  Speaker 4: I care about but maybe I don't care about.
• 29:27
  Speaker 5: I might go through them once a month or you
• 29:29
  Speaker 5: know whatever, and whenever I have some free time. Like
• 29:33
  Speaker 5: you look at that as now I've got every capture
• 29:36
  Speaker 5: costs me power, but I only have one the cost
• 29:39
  Speaker 5: of one upload.
• 29:40
  Speaker 4: Yeah, so that's.
• 29:41
  Speaker 5: Really where the savings comes from. Is you know, I'm
• 29:44
  Speaker 5: willing to spend power on something that I want in
• 29:48
  Speaker 5: the way that I want it, but I'm not willing
• 29:51
  Speaker 5: to spend as much power to get the stuff I
• 29:53
  Speaker 5: may or may not want, but I still want it.
• 29:54
  Speaker 5: It's worth something to me, it's just not worth as
• 29:56
  Speaker 5: much as these other things are.
• 29:58
  Speaker 2: Yeah, and what about this, Uh, correct me if I'm wrong,
• 30:01
  Speaker 2: But I think the new, the new, the brand new
• 30:04
  Speaker 2: camera's coming out.
• 30:04
  Speaker 4: Yeah.
• 30:05
  Speaker 2: I think it's called buck shot right where there's some
• 30:08
  Speaker 2: specific stuff for a buck Can you walk me through
• 30:11
  Speaker 2: what that looks like?
• 30:11
  Speaker 4: Yeah? Sure?
• 30:12
  Speaker 5: Uh so, so Buckshot essentially uses everything we've been talking
• 30:14
  Speaker 5: about today, and it just makes it easier for you
• 30:18
  Speaker 5: to view those pictures of the buck.
• 30:19
  Speaker 4: So how many of us sit.
• 30:21
  Speaker 5: There on our phone, you know, and they're like, oh, man,
• 30:23
  Speaker 5: look at that guy back there in the shadow, and
• 30:25
  Speaker 5: we're pinching and zooming and.
• 30:26
  Speaker 4: Like turn this out.
• 30:27
  Speaker 5: You know, we want to see every angle or every
• 30:29
  Speaker 5: color contrast that we can. Buckshot does that for you.
• 30:33
  Speaker 5: So now with just a simple tap gesture in the app,
• 30:35
  Speaker 5: you're like, oh, is that a buck back there? And
• 30:37
  Speaker 5: there's a chip in the right corner that says it's
• 30:39
  Speaker 5: Buckshot enabled. What that means is is that we have
• 30:43
  Speaker 5: we see we've seen something in the image, and that
• 30:46
  Speaker 5: something is something it's a it's a mature buck, or
• 30:48
  Speaker 5: it's a buck, and then when you tap and hold
• 30:51
  Speaker 5: on that image, it will essentially zoom it in and
• 30:53
  Speaker 5: give you a secondary image of that buck, you know,
• 30:57
  Speaker 5: from where we drew the box.
• 30:58
  Speaker 4: Around it, you know, where that's like a modeloring the
• 31:02
  Speaker 4: box around it.
• 31:03
  Speaker 3: And so is that just a zoom in or is
• 31:04
  Speaker 3: it also HD two?
• 31:06
  Speaker 4: Right?
• 31:06
  Speaker 3: Is that correct?
• 31:08
  Speaker 5: Yeah, the thumbnail, not the thumbnail, but the HD image.
• 31:13
  Speaker 5: It uploads an HD image automatically. If you have bush
• 31:16
  Speaker 5: that turned on. Sorry, I skipped that. Thank you, thank
• 31:19
  Speaker 5: you for bringing it back around of course.
• 31:22
  Speaker 2: So all right, so we can do a lot of
• 31:25
  Speaker 2: this onboard decision making, which is going to be the
• 31:29
  Speaker 2: camera makes the decision in the moment and capture something
• 31:32
  Speaker 2: or delivers something, and then there's some more of that
• 31:35
  Speaker 2: happening in the app. Is there any other is are
• 31:38
  Speaker 2: there any other sides of this AI equation when it
• 31:41
  Speaker 2: comes to sorting things, analyzing things, finding patterns when I'm
• 31:46
  Speaker 2: actually doing it in the app and trying to understand
• 31:48
  Speaker 2: this stuff?
• 31:49
  Speaker 3: What's that next level?
• 31:53
  Speaker 5: Sorry you broke up right there at the end, But
• 31:54
  Speaker 5: when you say next level, you mean like, what's the
• 31:56
  Speaker 5: next level of AI in the app?
• 31:58
  Speaker 4: Right?
• 31:59
  Speaker 1: Yeah?
• 31:59
  Speaker 2: So how you know this was which We've talked a
• 32:02
  Speaker 2: lot about what the camera is doing on board in
• 32:04
  Speaker 2: the moment. Now what about how does the app use
• 32:07
  Speaker 2: this technology for me to better understand what I'm looking
• 32:10
  Speaker 2: at and filter and and parse through these things?
• 32:13
  Speaker 4: Sure? Yeah, so, uh so obviously there's the what we
• 32:16
  Speaker 4: talked about.
• 32:17
  Speaker 5: Majority of camera companies now give you free access.
• 32:20
  Speaker 4: It's just through the app.
• 32:21
  Speaker 5: Is as I can filter by tags that have automatically
• 32:25
  Speaker 5: been placed on the image for me. You know, and
• 32:28
  Speaker 5: then the next level like what might be behind a paywall?
• 32:32
  Speaker 5: You know, is there some predictive stuff too, you know,
• 32:34
  Speaker 5: like we we do like multi doesn't activity charting, which
• 32:38
  Speaker 5: I think might be free to quote me on that.
• 32:41
  Speaker 5: I can probably dig through my notes if it's something
• 32:43
  Speaker 5: you want to know. But we do activity charting, which
• 32:46
  Speaker 5: essentially says, for this specific camera, what were where was
• 32:50
  Speaker 5: the peak activity you know, throughout the day. You know
• 32:56
  Speaker 5: you can modify that, you can say what was the
• 32:58
  Speaker 5: peak activity in the month or the week? I think
• 32:59
  Speaker 5: you like there are there are triggers on that that
• 33:02
  Speaker 5: that allow you to kind of you know, set what
• 33:05
  Speaker 5: you want. But yes, we we do activity charting and
• 33:08
  Speaker 5: then we do predictions too, and those predictions that those
• 33:12
  Speaker 5: are kind of the cool one. And I'm not really
• 33:14
  Speaker 5: sure how much the secret sauce I should let out there.
• 33:16
  Speaker 5: But you know, it's not just pictures from your camera mark,
• 33:21
  Speaker 5: It's it's aggregated data that we have access to that
• 33:26
  Speaker 5: you know, we don't we don't share with anybody else.
• 33:29
  Speaker 4: I mean, let me clear that up.
• 33:30
  Speaker 5: You know, it's like it is our data, it is
• 33:33
  Speaker 5: it's your data that that you know in that that
• 33:37
  Speaker 5: we we have access to and we use, you know,
• 33:40
  Speaker 5: a lot of different things like that to put into
• 33:43
  Speaker 5: our predictive models that essentially, you know, when you look
• 33:46
  Speaker 5: at it, it'll it's looking at not only activity from
• 33:49
  Speaker 5: your devices. It may be looking at activity from devices
• 33:53
  Speaker 5: in the you know, within thirty miles of your your
• 33:57
  Speaker 5: device or something like that, or maybe one hundred miles
• 33:59
  Speaker 5: of your device, you know, and it's factoring in what
• 34:02
  Speaker 5: your property is doing as well as you know what
• 34:04
  Speaker 5: surrounding properties might be doing.
• 34:06
  Speaker 4: You know, like.
• 34:08
  Speaker 5: Might might be an indicator of like a small second
• 34:11
  Speaker 5: rut you know, or a small like follow up rut
• 34:13
  Speaker 5: at the end that you know, maybe you didn't even
• 34:15
  Speaker 5: know anything about, but you know, a I picked up
• 34:18
  Speaker 5: on this weird you know one seven day peak or
• 34:22
  Speaker 5: three day peak or something, you know, and it's there,
• 34:24
  Speaker 5: been there for the last two years.
• 34:27
  Speaker 4: You know that that's the I guess.
• 34:29
  Speaker 5: That would be the other piece to put in that
• 34:30
  Speaker 5: predicted piece is that you know, you can go back
• 34:33
  Speaker 5: and look year over year in both activity and our
• 34:37
  Speaker 5: prediction models are looking at that. But in activity, you know,
• 34:40
  Speaker 5: if you were curious, like how you know my landing
• 34:43
  Speaker 5: strip food plot, how much activity was in it last
• 34:48
  Speaker 5: year or the year before, compared to how much I've.
• 34:50
  Speaker 4: I have now.
• 34:51
  Speaker 5: You know, maybe I don't know, maybe you changed the
• 34:54
  Speaker 5: change the food plot planning, or you know, there's a
• 34:56
  Speaker 5: clear cut now that runs right up the side of
• 34:58
  Speaker 5: it where it used to be, you know, a mature
• 35:01
  Speaker 5: stand of hardwoods, you know, mass producing stuff.
• 35:05
  Speaker 4: So you know, like there's a lot of stuff that
• 35:08
  Speaker 4: goes into it. I mean you can. I don't think
• 35:10
  Speaker 4: we could touch on everything. Those models are complex.
• 35:15
  Speaker 5: You know, there's somebody's way smarter than me out there
• 35:17
  Speaker 5: thinking like what if we added that into it?
• 35:20
  Speaker 4: What if we did it like that?
• 35:21
  Speaker 2: Yeah, So so give me some more of like the
• 35:26
  Speaker 2: how it's, how it's actually factored into, like the way
• 35:28
  Speaker 2: you use it in your own hunts when you're when
• 35:31
  Speaker 2: you're looking at this, and I guess one of the
• 35:33
  Speaker 2: things I'm also wondering wondering about is are you able
• 35:35
  Speaker 2: to it all separate between just your data versus the
• 35:40
  Speaker 2: collective data. Is there a way to say, hey, tell
• 35:42
  Speaker 2: me predictions based on just what my cameras have said
• 35:45
  Speaker 2: over the last six months or six years, and then
• 35:49
  Speaker 2: can I also then expand that and say, hey, give
• 35:52
  Speaker 2: me a broader sense of things based on everything else
• 35:56
  Speaker 2: you mentioned that secret sauce.
• 35:58
  Speaker 5: Yeah, I can so. Like A good example is is
• 36:01
  Speaker 5: I guess, like two years ago, I hunt here in Alabama,
• 36:06
  Speaker 5: you know south, I guess South Alabama, maybe Colin just
• 36:09
  Speaker 5: south Montgomery. But anyway, what you see a lot of
• 36:12
  Speaker 5: times is or what I what I like to do
• 36:15
  Speaker 5: is all leading up starting now starting you know, when
• 36:18
  Speaker 5: the velvet start, this horn start to develop, and you
• 36:20
  Speaker 5: know early on in the season, you can kind of
• 36:22
  Speaker 5: tell like, oh, that guy's probably gonna be a good
• 36:24
  Speaker 5: one right there, you know, And what you're doing as
• 36:26
  Speaker 5: you watch him grow is you're kind of.
• 36:28
  Speaker 4: You're putting him in a box. You know.
• 36:30
  Speaker 5: Does he like it when it's hot? Does he like
• 36:32
  Speaker 5: it when it's cold? Does he like it when the
• 36:34
  Speaker 5: wind blows out of the south or wind blows out
• 36:36
  Speaker 5: of the Norse, like like, all of that stuff is
• 36:39
  Speaker 5: stuff that you and I as hunters, we we.
• 36:43
  Speaker 4: We have to we had to do that ourselves. You know,
• 36:47
  Speaker 4: it was probably three years ago, two years ago.
• 36:49
  Speaker 5: I had to take notes and note, you know, hey,
• 36:51
  Speaker 5: he shows up on the north wind when the temperatures
• 36:55
  Speaker 5: you know, between fifty and seventy degrees.
• 36:57
  Speaker 4: Or something like that.
• 36:58
  Speaker 5: You know, but all of that now can you can
• 37:01
  Speaker 5: you can pretty much look at that in in your
• 37:04
  Speaker 5: you can go back and look at statistics for that.
• 37:07
  Speaker 5: So in your like activities you can go in and
• 37:10
  Speaker 5: look for that particular device. You can't get down to
• 37:13
  Speaker 5: the deer though, Like what it what AI is doing
• 37:16
  Speaker 5: for us today is giving us We're not looking at this,
• 37:21
  Speaker 5: We're not looking at say, twenty four hours worth of data.
• 37:24
  Speaker 4: We're we're looking at two hours worth of data.
• 37:27
  Speaker 5: Where AI said the deer are all moving in the afternoon,
• 37:31
  Speaker 5: or maybe they're moving from nine to twelve, you know,
• 37:33
  Speaker 5: which is a weird you know movement there's a ton
• 37:36
  Speaker 5: of pressure in the area or something. Don't move in
• 37:37
  Speaker 5: the mornings and afternoons and move middle of the day
• 37:39
  Speaker 5: or something. But that's what AI does for you to day.
• 37:41
  Speaker 5: I'll tell you that, like in that spot, your deer
• 37:45
  Speaker 5: like to move between.
• 37:46
  Speaker 4: You know, I don't know, eleven and twelve.
• 37:49
  Speaker 5: And so and then you can you can you can
• 37:52
  Speaker 5: feather a temperature piece on top of it. You can
• 37:54
  Speaker 5: feather you know, the wind. Like all of that is
• 37:59
  Speaker 5: available in the app for you to kind of all
• 38:04
  Speaker 5: stick it all together somehow, So it's it's still somewhat manual.
• 38:08
  Speaker 5: Like to what you're asking, you know, is can I
• 38:12
  Speaker 5: look at it like a specific deer or deer movement? No,
• 38:15
  Speaker 5: we can't do that yet, not very well anyway. You know,
• 38:20
  Speaker 5: that still involves human intervention. But what we can look
• 38:23
  Speaker 5: at is is site specifics. You know, what was the wind,
• 38:26
  Speaker 5: what was the temperature at that particular site? You know,
• 38:30
  Speaker 5: I mean, you know how it is. You might have
• 38:33
  Speaker 5: ten food plots, you know, and two of them are
• 38:35
  Speaker 5: going to produce almost every year. You know, there's something
• 38:37
  Speaker 5: about it where those maturity are just like to move
• 38:40
  Speaker 5: through there that piece we can do now from the
• 38:44
  Speaker 5: aggregated data provided by A do you.
• 38:48
  Speaker 2: Feel that hunters or even more specifically just your customers,
• 38:53
  Speaker 2: what's like the adoption rate of using all this stuff?
• 38:56
  Speaker 2: Like are people Do people get it and are doing
• 39:00
  Speaker 2: all these things you're talking about and enjoying these benefits
• 39:02
  Speaker 2: or has there been a lag or folks not understanding it?
• 39:05
  Speaker 2: Or do you get what I'm what I'm.
• 39:08
  Speaker 5: Yeah, I do, Yeah, And that's kind of a that's
• 39:11
  Speaker 5: a tough one. I think it plays into that confidence
• 39:14
  Speaker 5: thing more than anything. From what I've seen sitting on
• 39:16
  Speaker 5: the I guess sitting from the outside looking in, because
• 39:21
  Speaker 5: that's that's what we listen to. Our customers. Our customers
• 39:24
  Speaker 5: drive a lot of our roadmap. You know, we're not
• 39:26
  Speaker 5: just sitting back here as you know some guy that
• 39:30
  Speaker 5: hunts twenty four to seven. You know, like in fact,
• 39:33
  Speaker 5: we're lucky to get to go out sometimes. But you know,
• 39:36
  Speaker 5: we're not thinking like I like it, so they'll like it.
• 39:39
  Speaker 5: You know, we're asking our customers a lot of questions.
• 39:42
  Speaker 5: You know, anytime we can interact with the customer, that's
• 39:43
  Speaker 5: what we're doing. So from from those customer interactions, I
• 39:47
  Speaker 5: think the confidence wanes a little bit. And you know,
• 39:50
  Speaker 5: like how good is that technology that we've been talking
• 39:54
  Speaker 5: about at AI or object detection technology and then.
• 39:57
  Speaker 4: The aggregation piece.
• 39:58
  Speaker 5: You know, I think the confidence there, and I think
• 40:01
  Speaker 5: with that confidence, I think that's kind of what keeps people.
• 40:06
  Speaker 4: You know, that's the entry, that's the barrier of entry.
• 40:08
  Speaker 5: You know, Am I confident enough in it to spend
• 40:10
  Speaker 5: time to learn it, to spend time to use it?
• 40:12
  Speaker 4: You know? And I think I think.
• 40:14
  Speaker 5: Our goal as you know Moultrie are Yeah, our goal
• 40:20
  Speaker 5: as Moultrie and what what our customers deserve from us,
• 40:24
  Speaker 5: is for us to just give that to them, you know,
• 40:27
  Speaker 5: they shouldn't really have to learn anything.
• 40:28
  Speaker 4: And I think that's the part that's coming.
• 40:31
  Speaker 5: I think, you know, we've been nailing down technology, making
• 40:33
  Speaker 5: sure everything works like it's supposed to, and now we're
• 40:36
  Speaker 5: trying to figure out how to bolt it, bolt it
• 40:38
  Speaker 5: all together, you know, like all.
• 40:39
  Speaker 4: The components to build the house are here.
• 40:41
  Speaker 5: Now we just need to raise some you know, raise
• 40:44
  Speaker 5: some walls and put some nails in it.
• 40:45
  Speaker 4: And I think that's what's happening now, you know, is
• 40:48
  Speaker 4: we're putting those walls up.
• 40:49
  Speaker 5: We're nailing everything together now to deliver something you know
• 40:53
  Speaker 5: that that's going to be extremely useful. It's going to
• 40:57
  Speaker 5: be easy for the customer to use, you know, it's
• 40:59
  Speaker 5: not going to be is difficult, and I use difficult loosely.
• 41:03
  Speaker 5: It's not difficult now, it's just kind of you know,
• 41:06
  Speaker 5: you do have to poke around in the app.
• 41:07
  Speaker 4: You have to use your time to do that, and
• 41:09
  Speaker 4: you could be honting or doing something else. You might
• 41:11
  Speaker 4: each other.
• 41:12
  Speaker 2: What's for someone who is hesitant about this stuff? What
• 41:18
  Speaker 2: would be your entry way suggestion? So for somebody who said, ah,
• 41:22
  Speaker 2: I don't know about this. I'm sticking to what I know,
• 41:24
  Speaker 2: and they have not gone down this road yet. What
• 41:26
  Speaker 2: would be that first thing you would suggest, like, hey,
• 41:29
  Speaker 2: at least try this filters the filters, okay, at.
• 41:33
  Speaker 4: Least try every day when you go into your gallery.
• 41:37
  Speaker 5: Don't just thumb you know what I mean, like say, okay, okay,
• 41:41
  Speaker 5: I'm off work.
• 41:42
  Speaker 4: I got off work at five.
• 41:43
  Speaker 5: I'm gonna sait in the easy chair, you know, for
• 41:46
  Speaker 5: thirty minutes an hour this afternoon. When you're doing that
• 41:49
  Speaker 5: and you're looking through your pictures, filter, say show me
• 41:51
  Speaker 5: all the Bucks. You know, go in there and set
• 41:54
  Speaker 5: a filter, just a single just a single filter, don't
• 41:57
  Speaker 5: try and get too crazy, and show me Bucks, those turkeys,
• 42:01
  Speaker 5: you know, and then see if it works. You know,
• 42:04
  Speaker 5: just just hone into what you want, you know, if
• 42:06
  Speaker 5: you want to see Bucks, ask the app to show
• 42:09
  Speaker 5: you just Bucks, and then look through that and say
• 42:12
  Speaker 5: it's pretty turned good.
• 42:13
  Speaker 2: You know.
• 42:13
  Speaker 5: Yeah, it might have missed one or two, but it
• 42:16
  Speaker 5: was it was. I got enough information out of this
• 42:18
  Speaker 5: to say that now.
• 42:19
  Speaker 4: I don't need to.
• 42:20
  Speaker 5: I don't I don't have to if I don't want
• 42:22
  Speaker 5: to thumb through all of those pictures to just see bucks,
• 42:26
  Speaker 5: you know, or I don't know, say you're just in
• 42:28
  Speaker 5: for activity, you know, taking taking your.
• 42:30
  Speaker 4: Son, your daughter hunting or something like that.
• 42:32
  Speaker 5: Taking a just to just any kid hunting or some
• 42:35
  Speaker 5: nine hunter. Maybe you know you're introducing them. You just
• 42:37
  Speaker 5: want them to see deer, you know. Go to go
• 42:40
  Speaker 5: to activity in Maultri app. It's a free thing. It's
• 42:43
  Speaker 5: a free free feature. Go to activity and look and
• 42:46
  Speaker 5: say when when should I go hunting? And say, well,
• 42:49
  Speaker 5: it tells me I should go in the morning. I
• 42:51
  Speaker 5: know I should go in the morning. And then if
• 42:53
  Speaker 5: you got multiple food plot options or multiple camera options
• 42:56
  Speaker 5: camera sets, you know, go look at one or two
• 42:59
  Speaker 5: of those that you think are good good and kind
• 43:01
  Speaker 5: of see what the app tells you. You know, yeah, okay,
• 43:03
  Speaker 5: it definitely says that there's a lot of activity at this.
• 43:07
  Speaker 4: Camera set from ten to twelve.
• 43:10
  Speaker 5: And I know that's true because I've been hunting that
• 43:12
  Speaker 5: camera set for like the last four or five years.
• 43:14
  Speaker 4: You know, like I kind of know when to go.
• 43:16
  Speaker 4: But you know, we know our properties, we know our sets,
• 43:19
  Speaker 4: we know our cameras bys the time.
• 43:20
  Speaker 5: You know, we know a lot about how the animals
• 43:24
  Speaker 5: generally move about our farms and properties we have access to.
• 43:28
  Speaker 4: So you can use that to build your confidence.
• 43:31
  Speaker 5: You know, I recommend that Actually that's what That's what
• 43:35
  Speaker 5: gives you that confidence level to say, you know, like,
• 43:37
  Speaker 5: I feel confident I can go in here and I
• 43:39
  Speaker 5: can filter for bucks because I just want to see
• 43:41
  Speaker 5: bucks right now, I see those bucks, or I go
• 43:45
  Speaker 5: to the activity charting and I see that, you know,
• 43:48
  Speaker 5: these are my three favorite spots, and the AI agrees
• 43:52
  Speaker 5: with with the way that I like to hunt my
• 43:54
  Speaker 5: three Fravoran spots.
• 43:55
  Speaker 4: You know that that gives you the confidence to say.
• 43:57
  Speaker 5: I really don't I really hardly ever hunted on this
• 44:00
  Speaker 5: this south end, you know much. I think I'm gonna
• 44:04
  Speaker 5: see I'm gonna put a couple cameras out on that
• 44:05
  Speaker 5: south end and I'm gonna see what see what the
• 44:08
  Speaker 5: app what Moultrie tells me about the south end, and
• 44:11
  Speaker 5: I might have to go try it out. You know,
• 44:14
  Speaker 5: now that you've kind of used you know, then now that.
• 44:18
  Speaker 4: You've used your you're one.
• 44:21
  Speaker 5: Hundred percent confidence of I've hunted this place for that
• 44:25
  Speaker 5: long and then look looked at the AI and the
• 44:27
  Speaker 5: AI essentially agrees with you.
• 44:29
  Speaker 2: Yeah, so you're kind of using your You're able to
• 44:31
  Speaker 2: confirm the AI with the place you have experienced, right,
• 44:35
  Speaker 2: which gives you confidence in the tool to then go
• 44:37
  Speaker 2: put it in a place you don't have personal experience,
• 44:39
  Speaker 2: and then you can get an assessment of that with
• 44:43
  Speaker 2: the camera, and now you can trust it because you
• 44:45
  Speaker 2: were able to kind of fact check the camera, right.
• 44:48
  Speaker 5: Yeah, I mean there's some like ancillary like things that
• 44:53
  Speaker 5: you get from the AI too, you know, like all
• 44:55
  Speaker 5: of that, especially behind a paywall, but like we do
• 44:57
  Speaker 5: like buck to dough ratios, we can now we can
• 45:00
  Speaker 5: now do that from the.
• 45:03
  Speaker 4: Object detection stuff. There might be a couple others that.
• 45:05
  Speaker 5: I'm missing, but you know, like things that just like
• 45:08
  Speaker 5: regular guys like me or you, you know, I may
• 45:10
  Speaker 5: not have I mean I do spend time doing some
• 45:13
  Speaker 5: cameras I raised, but you know a lot of people don't,
• 45:15
  Speaker 5: you know, so they don't really know what their buckt
• 45:17
  Speaker 5: to dough ratios are or what their fawning success was
• 45:21
  Speaker 5: or something like that. So you know, like a lot
• 45:22
  Speaker 5: of that. Now can we can use AI to kind
• 45:25
  Speaker 5: of hit around it. You know, it may not be
• 45:28
  Speaker 5: as exact as me or you annitating images, but it
• 45:33
  Speaker 5: definitely gives you a good feel of should I take
• 45:36
  Speaker 5: those or you know, like am I healthy?
• 45:40
  Speaker 4: Is my property healthy? Or something?
• 45:52
  Speaker 2: How far away are we from, you know, just broadly
• 45:55
  Speaker 2: as an industry, how far away do you think we
• 45:59
  Speaker 2: are from being able to do you know, accurate antler scoring, age,
• 46:06
  Speaker 2: you know, estimation, all that kind of stuff.
• 46:08
  Speaker 4: Is that?
• 46:08
  Speaker 3: Is that pretty close?
• 46:10
  Speaker 4: I think it's pretty close.
• 46:12
  Speaker 5: I mean, you know, I think everybody's working on that now,
• 46:15
  Speaker 5: you know, I think we all know that, you know,
• 46:19
  Speaker 5: out out of the handful of camera competitors that are
• 46:22
  Speaker 5: out there now, I think that market. I think everybody
• 46:24
  Speaker 5: in that market's trying to solve that. I don't know how.
• 46:28
  Speaker 4: Close it is.
• 46:29
  Speaker 5: I think that that would be a super hard, you
• 46:31
  Speaker 5: know thing to put a date or even a timeframe
• 46:34
  Speaker 5: in are on.
• 46:36
  Speaker 4: But I definitely think everyone is working on it.
• 46:40
  Speaker 5: Everyone's trying to get that, and I think we're closer
• 46:43
  Speaker 5: now than we ever have. I mean, you saw the
• 46:45
  Speaker 5: you saw AI go from I can see a deer
• 46:48
  Speaker 5: to I can see a dough or buck to I
• 46:50
  Speaker 5: can see a mature buck or an immature dough. And
• 46:53
  Speaker 5: I think that's kind of where our an immature book,
• 46:56
  Speaker 5: I think, right Now that's kind of where it gets
• 46:58
  Speaker 5: a little you know, little little cloudy, you know. And
• 47:03
  Speaker 5: I think that's that's the natural progression you're going to see.
• 47:06
  Speaker 5: You know, I could see deer first. Now I can
• 47:09
  Speaker 5: see enough to delineate between buck and doe. Now I
• 47:12
  Speaker 5: can see enough to delineate between mature buck and immature buck.
• 47:15
  Speaker 5: Next level is I can see enough to know that
• 47:18
  Speaker 5: that is the wide a you know, or you know,
• 47:21
  Speaker 5: and and that's the hardest part, you know, and you know,
• 47:24
  Speaker 5: like you still can't lose track of like we still
• 47:27
  Speaker 5: have to do our part, march, Like our part is
• 47:29
  Speaker 5: to make sure that our cameras in a good spot.
• 47:31
  Speaker 4: You know, it's it's it's as it's as stable for
• 47:35
  Speaker 4: the trigger as it can be.
• 47:36
  Speaker 5: It's it's tuned the way that the way that I
• 47:39
  Speaker 5: want it for that particular spot. Like we still have
• 47:40
  Speaker 5: to do our spot. Are are part of, you know,
• 47:43
  Speaker 5: getting it ready to go. But I and the better
• 47:47
  Speaker 5: we do it that the better.
• 47:50
  Speaker 4: The better. The answer to your question that you just
• 47:52
  Speaker 4: gave me is winter we going to be able to.
• 47:54
  Speaker 5: Know that that is a specific deer regardless of where
• 47:58
  Speaker 5: that camera is while it's on the north end or
• 48:00
  Speaker 5: the south end. If that deer crosses one of those
• 48:02
  Speaker 5: two cameras, because he's done, because.
• 48:05
  Speaker 4: I've seen him before, I will know that that's him.
• 48:08
  Speaker 5: I think the better we set up our cameras and
• 48:11
  Speaker 5: give these really smart engineers the opportunity, you know, the
• 48:15
  Speaker 5: best best leg to stand on.
• 48:17
  Speaker 4: You know that, I think we'll be there in no time.
• 48:19
  Speaker 5: But if you know that technology's ability to filter the
• 48:25
  Speaker 5: chaff out is not that it's not that good yet,
• 48:29
  Speaker 5: you know, maybe there'll be a breakthrough in that which
• 48:31
  Speaker 5: will speed up the delivery of Can I identify a
• 48:35
  Speaker 5: specific deer, our specific buck?
• 48:38
  Speaker 4: But you know, if we.
• 48:39
  Speaker 5: Don't get any better at that, then you know it
• 48:41
  Speaker 5: might might be some time where the object dictation itself,
• 48:45
  Speaker 5: it just has to see more.
• 48:47
  Speaker 4: We just have to give it more.
• 48:48
  Speaker 5: It might have to see that same deer ten times
• 48:51
  Speaker 5: to get ten different angles of it and then I
• 48:52
  Speaker 5: know that that's that deer.
• 48:54
  Speaker 2: Yeah, that seems like I mean from the little I
• 48:58
  Speaker 2: understand about how these others or trained, right, it's just
• 49:02
  Speaker 2: for a long time there's been have been in putting
• 49:04
  Speaker 2: more and more training data. And the more training data
• 49:06
  Speaker 2: you put in there, and the more compute more power
• 49:08
  Speaker 2: you put behind.
• 49:09
  Speaker 3: It, the more the more these things are able to do.
• 49:12
  Speaker 2: So it seems like that would that would hold true
• 49:13
  Speaker 2: with something like you know, image detection and analysis, right, yeah,
• 49:19
  Speaker 2: do you know of any interesting ways or have you
• 49:22
  Speaker 2: played around with any interesting ways yet yourself of using
• 49:28
  Speaker 2: AI from other providers to kind of take another step
• 49:33
  Speaker 2: with trail camera data or hunting dead, like, is there
• 49:35
  Speaker 2: some way to integrate with chat GPT or Google Gemini
• 49:39
  Speaker 2: or anything like that to to take your analysis further
• 49:43
  Speaker 2: or to I don't know, maybe there'd be some way
• 49:45
  Speaker 2: to upload your photos that your that your camera has
• 49:49
  Speaker 2: told you were you know, of interest, and then you
• 49:51
  Speaker 2: go and could I upload those then to chat GPT
• 49:55
  Speaker 2: and have them do a next level thing?
• 49:57
  Speaker 3: Like is there some way to to go further.
• 50:02
  Speaker 5: So on the stuff that's outside of like camera apps,
• 50:06
  Speaker 5: you know, like GPT vision or something.
• 50:09
  Speaker 4: I think that's chat GPT's.
• 50:11
  Speaker 5: Version of of you know, like an object detection kind
• 50:15
  Speaker 5: of you know vision model.
• 50:18
  Speaker 4: I mean, really we have messed with uploading images to those.
• 50:24
  Speaker 4: What we see a.
• 50:25
  Speaker 5: Lot of times is is that those models aren't specifically
• 50:27
  Speaker 5: trained for deer or for turkey. They're trained for animal
• 50:33
  Speaker 5: or people or car or something like that.
• 50:36
  Speaker 4: So what we generally.
• 50:37
  Speaker 5: See is that that that type of stuff can tell
• 50:39
  Speaker 5: you that there's a deer in it, yeah maybe, or
• 50:42
  Speaker 5: maybe it can tell you if it's a deer in it,
• 50:44
  Speaker 5: but it could definitely tell you there's an animal.
• 50:45
  Speaker 4: There, you know, or you might call a kyotea dog
• 50:48
  Speaker 4: or something.
• 50:49
  Speaker 5: But what we see is that our models are much
• 50:51
  Speaker 5: more accurate because we actually train them on the same
• 50:53
  Speaker 5: number of images I would assume as their models are trained.
• 50:57
  Speaker 5: But what our model, what our training say do is
• 51:00
  Speaker 5: it's all deer. It's ten thousand pictures of deer, not
• 51:04
  Speaker 5: you know, a thousand of a dog, a thousand of
• 51:06
  Speaker 5: a person thousand you know. And I don't know how
• 51:08
  Speaker 5: they do theirs. I'm you know, I'm just kind of
• 51:11
  Speaker 5: riffing on that. But we see that our models are
• 51:13
  Speaker 5: generally speaking more accurate for what we want to see
• 51:17
  Speaker 5: as hunters because we train our models to show what
• 51:21
  Speaker 5: we want.
• 51:21
  Speaker 4: To see as hunters. As far as like data aggregation goes,
• 51:25
  Speaker 4: you know, like.
• 51:27
  Speaker 5: I mean, you can upload images into the Moultrie app now,
• 51:30
  Speaker 5: so like if you want to use all of the
• 51:32
  Speaker 5: tools that I've been referencing from Moultrie. You know, like
• 51:35
  Speaker 5: we allow you on the on the free side to
• 51:38
  Speaker 5: upload a certain number of images to our to our
• 51:41
  Speaker 5: app so that you can use our AI based tools.
• 51:45
  Speaker 5: It'll run through all of our tagging, you know, if
• 51:48
  Speaker 5: there is data with the image, we'll pull we'll scrub
• 51:51
  Speaker 5: it and pull it in. But but generally speaking like,
• 51:56
  Speaker 5: there's no, there's no. I don't know that there's anything
• 52:02
  Speaker 5: I have not experienced or experimented with anything that would
• 52:06
  Speaker 5: that I feel would give me anything more than we
• 52:09
  Speaker 5: already give you in the app, Like with the prediction models,
• 52:15
  Speaker 5: you know, the activity charting and stuff like that. I'm
• 52:18
  Speaker 5: sure you could take that to the anth degree with
• 52:21
  Speaker 5: some of these A platforms.
• 52:23
  Speaker 4: I mean, they're amazing. Yeah, it's amazing some of them
• 52:27
  Speaker 4: can do.
• 52:28
  Speaker 2: I've never seen this, so I'm assuming the answers no,
• 52:32
  Speaker 2: and probably nobody would actually want to do this, but
• 52:34
  Speaker 2: maybe a nerd like me would. Is there any kind
• 52:37
  Speaker 2: of export function within your guys' app in which I
• 52:42
  Speaker 2: could export the data.
• 52:45
  Speaker 3: So, for example, if I were to.
• 52:47
  Speaker 2: Say, all right, I've I've tagged every single time that
• 52:51
  Speaker 2: I've got a picture of the white A, We'll say right,
• 52:54
  Speaker 2: And so for three years, I've been tagging in my
• 52:57
  Speaker 2: in my app every time I see them, and you're
• 53:00
  Speaker 2: your cameras have got all this, you know, time and
• 53:04
  Speaker 2: date and wind and temperature, all that data is there, right,
• 53:06
  Speaker 2: So I'll collect on those photos and you guys do
• 53:10
  Speaker 2: a certain amount of analysis in the app on it.
• 53:13
  Speaker 3: But is there any way that I could export.
• 53:15
  Speaker 2: That data and then have like an Excel document that
• 53:18
  Speaker 2: shows two thousand data points that have all that information
• 53:22
  Speaker 2: tied to each one of those photos that then I
• 53:25
  Speaker 2: could then take and feed into chat, GPT or Gemini
• 53:29
  Speaker 2: and have it create for me, you know, further analysis
• 53:32
  Speaker 2: in the way I want to see it.
• 53:34
  Speaker 3: Is that technically possible.
• 53:37
  Speaker 4: Not that I'm aware of.
• 53:39
  Speaker 5: I think if it were, I think you and I
• 53:41
  Speaker 5: probably having a slightly different conversation about using some of
• 53:44
  Speaker 5: those outside AI platforms you know that have a lot
• 53:49
  Speaker 5: more compute than say, the device does, are even I mean,
• 53:53
  Speaker 5: our cloud servers have, you know, plenty of compute. If
• 53:56
  Speaker 5: we don't have enough, we get more. But you know,
• 53:58
  Speaker 5: like just their ability to crunch those numbers might be.
• 54:01
  Speaker 4: Better than what we can do. But not that I'm
• 54:04
  Speaker 4: aware of.
• 54:04
  Speaker 5: We there's nothing that says we can export, you know,
• 54:07
  Speaker 5: a whole bunch of images with the data that comes
• 54:09
  Speaker 5: with those images and put it into some CSV file
• 54:12
  Speaker 5: or something and uploaded to the you know, to a
• 54:15
  Speaker 5: third party man model interface or whatever.
• 54:21
  Speaker 2: I'm probably I'm probably the one in a million crazy
• 54:24
  Speaker 2: person that would be interested in doing that.
• 54:26
  Speaker 3: So I don't think.
• 54:27
  Speaker 4: We're doing well. We can do someone.
• 54:29
  Speaker 5: We do some of that for you mark to a
• 54:31
  Speaker 5: certain extent, you know, like when you when you go
• 54:33
  Speaker 5: look at our prediction model, like you can kind of
• 54:36
  Speaker 5: see things in our prediction model as you as you
• 54:39
  Speaker 5: pay over like the days and the hours of when
• 54:42
  Speaker 5: you might want to go hunting. You'll see that the
• 54:45
  Speaker 5: weather changes as you move it, the wind changes as
• 54:47
  Speaker 5: you move it, you know, and and then even at
• 54:51
  Speaker 5: even at the activity level, you know, like we're we're
• 54:54
  Speaker 5: we're putting some of those data points out there for
• 54:57
  Speaker 5: you to see. You know, like when when you look
• 54:59
  Speaker 5: at peak movie at a particular camera set, we're not
• 55:02
  Speaker 5: only showing you like, hey, there was a bunch of
• 55:04
  Speaker 5: deer that this camera took at this time and this time,
• 55:07
  Speaker 5: we're also showing you on other you know, a little
• 55:10
  Speaker 5: further down the scroll is what was the wind doing
• 55:13
  Speaker 5: whenever those peak moments were happening, you know, or what
• 55:17
  Speaker 5: was the temperature doing at those peak moments or something
• 55:20
  Speaker 5: like that. So again like really the only slice we're
• 55:23
  Speaker 5: missing is is that what dear was it that we
• 55:26
  Speaker 5: saw at in the morning and in the afternoon. Like,
• 55:30
  Speaker 5: if we can win, we break through that, it'll definitely be.
• 55:36
  Speaker 4: Amazing. Yeah, it'll be super cool.
• 55:39
  Speaker 2: So okay, So the the individual buck detection, the Antler
• 55:44
  Speaker 2: scoring age estimation, those are things that you mentioned that
• 55:49
  Speaker 2: probably everybody's working on to some degree, and that's coming
• 55:52
  Speaker 2: at some point. What else might we be able to
• 55:56
  Speaker 2: imagine in the future when it comes to how AI
• 55:59
  Speaker 2: might you know, integrate with this with this technology we're
• 56:03
  Speaker 2: using now, I don't know, like five years from now,
• 56:04
  Speaker 2: ten years from now, if you're just putting on your
• 56:07
  Speaker 2: your imagination hat and thinking about where this maybe could go,
• 56:10
  Speaker 2: what might be possible? Uh, what else might be on
• 56:14
  Speaker 2: the horizon?
• 56:16
  Speaker 4: I don't know.
• 56:17
  Speaker 5: If I don't know if my employer would like for
• 56:20
  Speaker 5: me to say too much about that, Mark, I.
• 56:22
  Speaker 4: Guess that's a good point, but I can What I.
• 56:26
  Speaker 5: Will say is that if you can think it, I
• 56:30
  Speaker 5: think in the future there will be an easier way
• 56:34
  Speaker 5: for it to be done. You know, the days of
• 56:38
  Speaker 5: looking at a picture and or let's say, looking at
• 56:42
  Speaker 5: a camera pan on a map and equating that to
• 56:46
  Speaker 5: what you see in the gallery by saying, you know,
• 56:49
  Speaker 5: I saw this big deer moved across this camera when
• 56:55
  Speaker 5: the when the temperature was fifty degrees and the wind
• 56:57
  Speaker 5: was out of the north, and barometric pressure was you know,
• 57:00
  Speaker 5: I don't know, twenty eight and all of that. You know,
• 57:03
  Speaker 5: like the days of that are not necessary anymore. I mean,
• 57:07
  Speaker 5: maybe I went out on a limb that'll break by
• 57:09
  Speaker 5: saying it's not necessary, I don't know, but I definitely
• 57:13
  Speaker 5: think that you know that is is here now. Just
• 57:18
  Speaker 5: tapping into.
• 57:18
  Speaker 4: It and learning how to use it in a way that.
• 57:20
  Speaker 5: You feel comfortable and confident with it is is kind
• 57:23
  Speaker 5: of the the mud we're stuck in right now.
• 57:27
  Speaker 3: All right.
• 57:28
  Speaker 2: So I'm going to ask you to repeat a little
• 57:30
  Speaker 2: bit of some stuff you've already said, but I want
• 57:32
  Speaker 2: to kind of tie a bow on this. If you
• 57:35
  Speaker 2: were to give me the three things that every deer
• 57:40
  Speaker 2: hunter should try to be using when it comes to
• 57:44
  Speaker 2: AI and their trail cameras, what would be like the
• 57:46
  Speaker 2: like the three things that hey, everybody, you should try this.
• 57:49
  Speaker 2: You should wrap your head around this. You should be
• 57:51
  Speaker 2: trying to take advantage of this. And if you can think,
• 57:53
  Speaker 2: I mean you can you can speak broadly like across brands,
• 57:56
  Speaker 2: or if there's some stuff that you need to you guys,
• 57:58
  Speaker 2: that's fine too. But I'd be curious to hear you know,
• 58:00
  Speaker 2: everybody out there, these are three things you really should
• 58:03
  Speaker 2: give a shot. What would those three things be.
• 58:07
  Speaker 5: Be filtering in the app Obviously that that is the
• 58:09
  Speaker 5: lowest level, and the thing that's that's really the tip
• 58:14
  Speaker 5: of the spear.
• 58:15
  Speaker 4: That's what's gonna set some confidence.
• 58:17
  Speaker 5: You know, that's gonna it's gonna it's essentially just confirming
• 58:21
  Speaker 5: what you see by using the filters.
• 58:22
  Speaker 4: You're already looking through your gallery. Look through your gallery,
• 58:26
  Speaker 4: go out, take a wild guess.
• 58:28
  Speaker 5: And say, you know, let me see if it sees
• 58:31
  Speaker 5: but the same bucks I just saw.
• 58:33
  Speaker 4: Do that. It's easy, it's not gonna take long.
• 58:36
  Speaker 5: It's definitely gonna give you some confidence in the technology.
• 58:40
  Speaker 5: The second piece is is probably using like those the
• 58:43
  Speaker 5: for multum I almost speak to Moultrial on this one.
• 58:45
  Speaker 5: The previous comment can be done across a lot of
• 58:49
  Speaker 5: different camera apps. Yep, this one the smart Capture OPPS
• 58:53
  Speaker 5: based stuff obviously based settings like that's gonna be from
• 58:56
  Speaker 5: what I know. It's a multie you know, you need
• 58:59
  Speaker 5: it's you need tumultuary we And that's the one that
• 59:02
  Speaker 5: saves you power, saves you a little bit of cell
• 59:04
  Speaker 5: you o their you know, data for your uploads and stuff.
• 59:06
  Speaker 4: But that's the one where.
• 59:08
  Speaker 5: You get what you want, when you wanted, how you
• 59:10
  Speaker 5: wanted essentially, so like you know, now that you've filtered
• 59:13
  Speaker 5: a little bit.
• 59:13
  Speaker 4: Got some confidence in it.
• 59:14
  Speaker 5: Next level is let me, let me tell my device
• 59:18
  Speaker 5: to just take videos or just take pictures whatever it is,
• 59:22
  Speaker 5: whatever whatever you're viewing media is, whether that's pictures or videos,
• 59:26
  Speaker 5: tell it.
• 59:26
  Speaker 4: I just want to.
• 59:27
  Speaker 5: See bucks in the form of a fifteen second video.
• 59:29
  Speaker 5: I want those to upload immediately, and I want everything
• 59:33
  Speaker 5: else that my camera takes to upload once a day.
• 59:37
  Speaker 5: Just I mean, you know, just try that for a
• 59:39
  Speaker 5: couple of days. You know, did it show me as
• 59:41
  Speaker 5: many bucks as I was seeing when I was uploading everything?
• 59:45
  Speaker 5: You know, now between eight and five pm, you know,
• 59:48
  Speaker 5: like I'm looking at my app and I'm not having
• 59:50
  Speaker 5: to look through the five or ten images that just
• 59:52
  Speaker 5: came in from three or four five different cameras. I'm
• 59:55
  Speaker 5: only seeing the bucks, so you know, like I'm not
• 59:57
  Speaker 5: having to weighe through everything, but it's X when I
• 1:00:00
  Speaker 5: get home at night. Maybe my cameras have started seeing
• 1:00:03
  Speaker 5: showing me the pictures of the of the pesky raccoons
• 1:00:06
  Speaker 5: that I have to look at every day, or you know,
• 1:00:09
  Speaker 5: the squirrels that are chewing holes in the top of
• 1:00:12
  Speaker 5: my feeter because I haven't put any corn in it
• 1:00:14
  Speaker 5: in a or something, you know you And then I
• 1:00:17
  Speaker 5: guess the third thing is is that is that data aggregation,
• 1:00:22
  Speaker 5: you know, and like it's really hard to say, go
• 1:00:24
  Speaker 5: look at activity or go look at prediction or something
• 1:00:28
  Speaker 5: like that. You know, I guess if you've done the
• 1:00:30
  Speaker 5: two things that I just mentioned, you're you're becoming a
• 1:00:34
  Speaker 5: super user to some extent. You know, like you're now
• 1:00:37
  Speaker 5: you're saving battery. You're using AI to upload what you
• 1:00:40
  Speaker 5: want when you want it. You're using the filters in
• 1:00:42
  Speaker 5: the gallery to only see what you want while you're
• 1:00:45
  Speaker 5: in there, regardless.
• 1:00:45
  Speaker 4: Of how they got uploaded.
• 1:00:47
  Speaker 5: Now you're going to say, like where do I need
• 1:00:49
  Speaker 5: to be and when do I need to be there?
• 1:00:51
  Speaker 4: On my property.
• 1:00:53
  Speaker 5: You know, can the app tell me that, Well, the
• 1:00:55
  Speaker 5: app maybe can't put a pin in that specific field
• 1:01:00
  Speaker 5: to hunt or something like that, but we can definitely
• 1:01:03
  Speaker 5: lead you in the direction of what time of day
• 1:01:06
  Speaker 5: if you're like me, like I seldom get the hunt
• 1:01:08
  Speaker 5: Monday through Friday. You know what I mean. We have jobs,
• 1:01:12
  Speaker 5: we do what we do. You know, So I am
• 1:01:14
  Speaker 5: in the field on Saturdays and Sundays come hell or eyewater,
• 1:01:18
  Speaker 5: and it doesn't matter if it's raining, it doesn't matter,
• 1:01:21
  Speaker 5: if it's freezing, sleep and snohing, it doesn't matter. Like
• 1:01:23
  Speaker 5: I'm out there, and when I'm out there, like I
• 1:01:26
  Speaker 5: just want to know what's the best time for me
• 1:01:28
  Speaker 5: to go get wet or go freeze or something like that,
• 1:01:31
  Speaker 5: you know, and those activity those activity things, like the
• 1:01:35
  Speaker 5: activity charting and the prediction models can start to help
• 1:01:38
  Speaker 5: you hone in on you know, like I'm a weekend warrior,
• 1:01:41
  Speaker 5: I only have two days to hunt. You know, maybe
• 1:01:44
  Speaker 5: maybe I only have two days to hunt two weeks
• 1:01:46
  Speaker 5: a month, you know, because my job or my family
• 1:01:48
  Speaker 5: keeps me from doing that. You know, that activity charting
• 1:01:51
  Speaker 5: and prediction model within Multrie's app will now increase your
• 1:01:56
  Speaker 5: odds on those two days twice a month that you
• 1:02:00
  Speaker 5: are able to hunt.
• 1:02:01
  Speaker 2: Yeah, all right, Well, i'd say those are three marching
• 1:02:05
  Speaker 2: orders that all of us can take out there in
• 1:02:08
  Speaker 2: the field and give a shot, and we'll come back
• 1:02:10
  Speaker 2: to you a year from now as AI power users.
• 1:02:13
  Speaker 3: And I'm going it back, all right, sweet.
• 1:02:16
  Speaker 2: Yes, I appreciate you, Darren, walking us through this, talking
• 1:02:21
  Speaker 2: through where things are right now with AI and maybe
• 1:02:23
  Speaker 2: where they're headed soon. I'm excited about that future when
• 1:02:27
  Speaker 2: all of my nerdy questions can be answered without me
• 1:02:33
  Speaker 2: having to spend seventeen hours a week analyzing spreadsheets.
• 1:02:36
  Speaker 4: Yeah.
• 1:02:37
  Speaker 5: No, I appreciate the opportunity to come on and talk
• 1:02:39
  Speaker 5: about it. I mean, I do think it's a maybe
• 1:02:42
  Speaker 5: underserves not the right word for it. Maybe underutilized is
• 1:02:45
  Speaker 5: more of the word, but it could be a little
• 1:02:47
  Speaker 5: of both. But you know, I do think it's important
• 1:02:49
  Speaker 5: for the customers to start to use the technologies this
• 1:02:52
  Speaker 5: out there. It'll definitely increase their odds, you know, in
• 1:02:56
  Speaker 5: the field, and it'll definitely make it more pleasurable to
• 1:03:00
  Speaker 5: look at pictures.
• 1:03:02
  Speaker 2: Love it all right, Darren, Well, thank you for being here,
• 1:03:04
  Speaker 2: and let's stay in touch as all of this continues
• 1:03:08
  Speaker 2: to evolve because if there's anything that I can say
• 1:03:10
  Speaker 2: about AI, it is that it is always.
• 1:03:12
  Speaker 3: Changing and growing, and the news.
• 1:03:15
  Speaker 2: Is always new, So I'm sure there'll be something interesting
• 1:03:18
  Speaker 2: to talk about next year or the year after that.
• 1:03:20
  Speaker 4: I hear you all right, sounds good, Mark.
• 1:03:23
  Speaker 2: Thanks all right, that's going to do it for us today.
• 1:03:27
  Speaker 2: Thank you for joining me for this conversation. If you'd
• 1:03:30
  Speaker 2: like to learn more about what Dared was speaking specifically
• 1:03:33
  Speaker 2: to with Moultrie, you can visit multiemobile dot com to
• 1:03:37
  Speaker 2: visit to explore their offerings and some of these specific
• 1:03:40
  Speaker 2: feature sets we've talked about Otherwise, Until next time, I
• 1:03:44
  Speaker 2: hope you get out there try some of these new
• 1:03:46
  Speaker 2: things on your cameras, regardless of what kind of camera
• 1:03:49
  Speaker 2: you're running, have some fun with it, do some tinkering,
• 1:03:52
  Speaker 2: do some experimenting. Until next time, stay wired to hunt