Ep 5: Will NeRFs change digital imaging?


Episode Categories:



Will something called NeRFs completely change the way that we look at visuals? I think it will.

So that's just one of the things that we're going to talk about today on Everyday AI. Your everyday podcast for the latest in AI news, tips and tricks can't wait. It's Monday morning and we have our rotating co host back. Brandon, good morning. How are you?

Brandon [00:00:31]:

I'm doing well. How are you doing?

Jordan Wilson [00:00:33]:

Doing fantastic. So let's get in. I'm going to quickly preview what we have going on for you at Everyday AI and then we're going to jump in.

So we're going to talk about NeRFs, what the heck that is, and how it affects you. We're going to talk about tech leaders mentioning AI an insane amount of times on their earning calls. And then we're going to talk about a Boston Globe story, something very interesting happening with chat GPT in Boston University.

Brandon and Jordan's Image and Video Background

So before we dive into those, Brandon, let's just throw this out there. NeRFs. So they're brand new. There's a new study that actually just came out yesterday, just talking about this different without getting too technical, because, again, Everyday AI is a podcast for the everyday person like you and me, people who aren't highly technical. So with these nerfs, they're neural radiance fields. It's using 3D imaging and all this crazy dorky stuff. But before we go in and talk about it, Brandon, you have a little bit of experience in photo and video, right?

Brandon [00:01:42]:

Yeah, I do have experience with both photography videography, working with obviously mostly DSLRs. I haven't had much work with Mirrorless, but yeah, I've had photo video work. I've worked with small capacity with drones, working with stabilizers, things like that. Some of the things that this just blows out of the water and can replicate like times ten. So it's crazy knowing how much manual work goes into setting up tracks, setting up drones, a little bit of background. And I did do like a short film, so I have been on a set and whatnot, but it's crazy that this thing is able to do what it can do based off of my prior knowledge. So it's amazing.

Jordan Wilson [00:02:20]:

Yeah. So I think it is worth mentioning. Yeah. Brandon has a photo video background. I even kind of do myself back when I was a multimedia journalist back in the day. And so Brandon and myself, I think, both personally realize the amount of work that it usually takes to get some high quality visuals. Not just high quality visuals, but Brandon mentioned two big things there. He mentioned drones and he mentioned stabilizers. Right. Being able to produce high quality photo and video on its own is pretty difficult. But being able to produce video that shows movement, people can spend their years or longer just trying to get decent at it, and it's extremely hard. Brandon, what are some of the challenges of getting video? And then we'll talk about the applications but what are some of the challenges of just getting video that shows movement?

Brandon [00:03:17]:

I mean, tons of things. You have like motion blur. That's one of the biggest things. Depending on the settings of the camera, if you're using manual, if you're using like sport or action mode that some of the maybe other cameras might have. Obviously, if you're talking about bigger equipment that's so heavy to carry, you're going to need a whole track to set up. Imagine like a railroad track for a train, but you have a huge device just following you along. Obviously in the cinematic field, you'll have even cars attached to cameras attached to cars that are passing by. So that's just so many. One, it's money. That's a huge thing to tackle. Do you have the money to be able to afford those kinds of things? And even if you're just using your own personal equipment, it's just getting that an image that doesn't look blurry, focusing on the right subject. There's just so many aspects to it, but those are just some of the few.

How NeRFs Will Change Image and Video Capture

Jordan Wilson [00:04:07]:

Yeah. And such an important thing. So time is big. It takes time to not just learn how to create good video with movement, but it's costly too. Right? So let's talk about this new technology. Nerfs. So not the gun, not the nerfs guns that little boys run around and shoot each other with growing up in the house with the foam darts, but neural radiance fields. Don't worry, it's super dorky technology. But the applications are amazing. They are amazing. And let's actually talk quick, Brandon, about what this means for the everyday person. Right? So not saying that you're going to jump in and learn to develop with this technology. That's not what we're talking about.

But how do you think, brandon, what do you think are some common day to day applications after just learning about this technology? It is pretty new. What are some everyday applications that you can see? And what does this mean, having this kind of technology? And we will share this if you're listening live on LinkedIn or anywhere else. Feel free to drop a comment if you want to know a little bit more. But we will drop this in our daily newsletter as well, so you can check out visual examples. Because if you're hearing this on a podcast or if you're watching this on the live stream, it's not going to make a lot of sense, but we will share it in our newsletter. But Brandon, just talk about some of the everyday applications that everyday people might be able to use this for, right? Yeah.

Brandon [00:05:40]:

So I guess for anyone listening, imagine a 3D image, but instead of it being some kind of rendering and it's in a sense animated, this is truly just like machine learning, taking an actual image and creating a real, actual 3D photo of this thing. So you're seeing what it actually looks like. The closest thing I can think of is maybe VR, but in image format. And so I'm thinking real application uses obviously, real estate is huge. So being able to tour real estate and even just architecture in general, so being able to tour a whole building, I mean, even before it's built as an example.

So I can't imagine, even if you're just looking if you're working even in a basic capacity, and you're not necessarily a contractor yourself. Maybe you're looking to build a new house as an example. You could probably create a whole NeRF of your house, if that's even the proper use of how we're going to call it, like a NeRF, and just see what it looks like.

And I'm even thinking this is probably further ahead, but the way we use AR, it's limited, but the way it's mostly pushed I know, on Amazon, like, see what this furniture looks like in your home. I mean, I can imagine. Those are obviously kind of like a little pixelized. They look a little bit animated. But imagine if this is somehow implemented into, like, phone AR use case. That would be amazing to be able to see what things look like right in front of you from a more realistic perspective. So that's what I'm thinking.

Jordan Wilson [00:07:05]:

Yeah, absolutely. I'd say the other thing is, it allows us to keep an eye on this technology. We'll continue to be talking about it probably on The AI Show and the AI newsletter, but I think it allows everyday people to start to create something, to create a story of their own, to give their if you're a small business founder. If you work in marketing on a small team, this gives you capabilities that would normally not be afforded to someone unless you had a six figure budget for video production. To get these kind of visuals, again, Brandon and I do have that background of shooting photo and video. You would need a high quality camera. You would need multiple people on a crew. You would need probably a 360 camera. You would need what Brandon talked about, like a stabilizer. You would probably need a drone shot. This combines all of those different technologies. So I just named three or four different cameras, multiple pieces of external equipment, multiple people.

The technology will get there. It's not perfect, but it is amazing with how new this technology is and the outputs. That's why I wanted to spend a good chunk of time this morning on it. Because for the everyday person, it can get your side hustle off the ground. It can grow your small business. If you're on the marketing team and no one else on your team knows about this, you can get in and you can create something that is going to blow everyone on your team away. So keep an eye on that, and we'll continue to talk about it. 

The Big Three Push for More AI

So let's real quick talk about the big three in AI. Right? You can make arguments for who the big three in AI are. Personally, I think it's Google, Microsoft, and Meta. So Meta is obviously Facebook instagram. WhatsApp? There might be one or two other that they bought recently, but those are kind of the big three. So we actually mentioned this in our newsletter, I think on Thursday or Friday, but we didn't get a chance to talk about it, and I think it's important. So there's a recent study that came out and analyzed the amount of times that AI was mentioned in quarterly earnings calls.

So big public companies like Meta, Google, Microsoft, they have quarterly earnings calls, and people have actually analyzed the amount of times that they've referenced AI. Artificial intelligence. And what these studies found is that it's exploded with this most recent earnings calls. Those three companies mentioned AI more than 170 times. And you all will recognize or realize that I did leave a pretty big company off that list. Apple, we've talked about that before. I do think Apple will come in and make some huge splashes in AI. But right now I wouldn't consider them top three players yet. I do think they will be. But let's get back brandon, hearing how many times these people mentioned AI on these calls. What do you think that means? What does that tell you?

Brandon [00:10:07]:

I think it's just obviously if you're kind of keeping up even with just daily AI news, you know, it's not just a trend. It's not just something that's going to fade like other things might have in the past. But I would say that this just makes the argument that this is something that smaller businesses and organizations should be paying attention to. If you haven't already, or if you've kind of seen it as like the cool party trick, it's definitely more than that. And obviously that's what our podcast is looking to do to help you realize that there's so much that you can apply to the use of AI. And so I think it's just solidifying that this is the future and it's already here.

And so for people that maybe may not know what's going on or may have an idea, hopefully you can kind of catch up and pick up what's going on so that you can make a small organization or a small company can make three times more movement than a large organization can in terms of structural changes just because of the way and the flexibility of them. And so for you to have the ability now to maybe take the step and match what bigger organizations like Google is doing right now is going to be huge. And so I think that's really important for people to pay attention to.

Jordan Wilson [00:11:15]:

Yeah, I think that's a great point, Brandon. Another something to talk about is I think this means something bigger for just the venture capital industry or the startup space, right. That's not everyone listening, but I do know that some of you listening probably are involved in that space. And I think that it's something important because anecdotally and looking at studies is it seems it's harder for startups to raise money if you do not have an AIP central to your company. So I do think that this is a trend that unfortunately will probably continue it will probably continue to be a little harder for smaller startups to raise money the same way they could a year ago, two years ago. If you do not have an AI central piece to your software, to your company, to your platform, whatever it may be, I don't think that's necessarily good for the industry, which might sound crazy, like, hey, Jordan, Brandon, you guys are talking on an AI podcast. Why do you think that's bad?

Because I don't think AI should be infused in every single startup because then it creates just this, I think a jumbled mess and groupthink. I think one of the reasons that AI has exploded over the last nine months is because not everyone was using it. And that allows people the creativity and the ability to think uniquely and to separate themselves from the pack. So if everyone is using AI, every single person, I think that actually dampens the likelihood for innovation. But that's a tangent for another day.

ChatGPT Being Used in The Classroom

So let's quickly talk about kind of our third big development. So, Boston Globe story came out over the weekend, talking about students working with faculty actually developed a policy for using artificial intelligence in the classroom, essentially saying it's open to use this technology. You just have to disclose when you use it. Seems like a pretty common sense policy to me. Brandon, what are your takes on this using AI in the classroom?

Brandon [00:13:32]:

Yeah, this kind of goes back to what we were just touching on, but I think it's good in a certain capacity. Of course you still want depending on how it's used, obviously you still want people to be able to learn on their own and develop their own thinking writing skills as an example. I know this is in a college setting, but even using it at a lower level, but I think it is a step in the right direction only in the sense that I think we should be honing in on those skills of using Chat GPT or other AI just to be able to have that knowledge.

So when you are going to look for something in your career, I wouldn't be surprised if that's like, in five years, like have a good knowledge of Chat GPT or be able to use this and this AI. And so I think it is something that should be introduced. I think college level is good. High school, I don't know. We could get to that. But yeah, I don't know. The only thing I'm thinking in the back of my head is how can we implement it in a way where you're still learning, but you're adapting and not just like having it do everything for you, but you're not realizing its capabilities for sure.

Jordan Wilson [00:14:43]:

I think this is important and we're going to continue to see a lot more of these stories and studies come out. My personal take is I think it should be required. To me, it's almost crazy, though. This is one of the first big stories that we've heard about official AI policies in the classroom. I think that students obviously have been using this technology. Students aren't dumb, they've been using it. If you're listening out there, if you know a college student or two in your life, ask them if they're using Chat GPT for their papers, their assignments. They all are. So you might as well not just make policies encouraging it.

But if I was a professor, I would make it a requirement. Instead of having a paper due every other week, there would be a paper every single day. There would be a paper due every single day. And you would do it in class. And you would have to leverage GPT just in order to get it done. And you would have to fact check it and make it well thought out and well researched. But I think that's the future of where learning is going. It's kind of think when I was in high school, it's kind of when the internet was first becoming very popular and we use the internet to help write our stories, right? Like, obviously sourced it and paraphrased it. So it's a little different. I get another conversation for another day because I'm going long on this and I do want to wrap this up.

Outro: AI Tip From Jordan & Win Premium ChatGPT

But it's great that we're talking about Chat GPT because as a reminder, everyday AI is giving away two year long premium subscriptions. So if you don't know what Chat GPT is, listen to some of our prior episodes, but you probably do. It's an amazing tool. The free tool is great. The paid subscription is definitely worth getting everything's much faster. You get some different capabilities that aren't in the free version as well. So make sure you go to youreverydayai.com, sign up for the daily newsletter that'll be in there.

And I'm also going to leave you, even though we're a minute over already, I'm going to leave you with one tip and trick. I'm going to try to throw out one thing that everyday people can do with AI to start integrating it into their daily lives. So I'm going to start there and I'm going to start with something very simple today, if you don't already, I would encourage you to open a Chat GPT account. So just go to OpenAI that's the owning the parent company open an account. And here's what I want you to do, something very specific. Instead of going to Google first, I want you to go to Chat GPT, see how it works. A couple of things to keep in mind. Chat GPT is limited in terms of freshness. So if you need something that's extremely, extremely relevant. So Chat GPT's kind of model only goes up through September 2021, keep that in mind.

But if you don't need the freshest of information, if you're just doing some research or something that's not timely, I encourage you to start in Chat GPT and talk to it as if it was someone on your team, as if it was a research assistant. We'll get into what that means later. But that's kind of the tip and trick for today. Don't use Google as much. Go in and use Chat GPT or you can use Google. Bard is kind of another one. Bing has Microsoft. Microsoft has their Bing chat. So instead of just using a normal Google search, use Chat GPT. Use one of these other kind of AI chat models. I think it's going to do wonders for your growth. So we went over that's. Okay? I think there are some important things to tackle today. Thank you so much for tuning in and we hope to have you back on the show every day. Again, this is Jordan and Brandon signing off for your everyday AI.

Gain Extra Insights With Our Newsletter

Sign up for our newsletter to get more in-depth content on AI