I think a bigger problem is how much computational resources NERF models need. From the paper,
"Our model, Mip-NeRF 360, and our 'mip-NeRF 360 + iNGP' baseline were all trained on 8 NVIDIA Tesla V100-SXM2-16GB GPUs."
Even with all that, it takes almost an hour to train. A developer would need to do this for every scene, which may make it unfeasible for indies, those who would benefit the most from NERF in gaming.
The better question is how many person hours would it take to model a scene like that with alternative approaches, and can a NeRF be used to create outputs that could be used in a game engine?
Thanks for the reply, I do think you're right. I'm also GenZ, and thinking about others does consume my mental state more than I'd like. Maybe just shutting off the content consumption for a while is a good idea.
Yea I do think I haven't found my niche yet, I tend to read about a lot of different things, and hard to focus on one thing specifically. Thanks for the advice.
Thanks, yea I originally did it just because I thought it would be a nice change of pace from my day to day, and a fun way to learn C++. The code isn't really good so I avoided pushing it to github.
Yea the artistic side of things might be good to try, I do like the idea of exploring toon shading more.
Please let me recommend SideFX Houdini as an artistic (yet technical) tool to learn:
https://www.sidefx.com
(perhaps would be a kind of therapy in itself - it was for me long ago)
> exploring toon shading
Sounds excellent!
And don't hesitate to publish your work.
Well, to make an (imperfect) analogy with programming tools - Houdini is like Emacs, while Blender is more like Visual Studio.
That is not to say that one is intrinsically better than the other, but for a technically-minded person (== any programmer in this case), the way Houdini works will make much more sense.
Learning either of these tools is not an easy task, but I think that starting with Houdini will feel more natural in cases like yours (SE background, interest in interactive graphics).
You can email me (see my profile here) for some more Houdini-talk!
Yea I think I am usually able to make a functional prototype, just creating something that's actually all the way finished and great is a challenge.
Might be a good idea to find someone to help make things into a longer form solution, but I'm fairly introverted/shy, so that's always a challenge, but good advice!
I agree, but it's hard sometimes, you know? I for some reason can't help but compare, even though it does no good.
Yea I am working as a professional backend software engineer. Started right after University.
For what I really want to do, I am not really sure anymore to be honest. There are jobs that I think would be very interesting and a nice change of pace, like Graphics Engineering, but I always end up thinking I'm too under qualified. I'm almost done my Masters in Computer Science, but still don't feel too qualified.
Okay so you are a professional and are getting paid to create, presumably, useful software.
In that case I would take some time, be loving and kind to yourself, enjoy life, and consider what you would really like to do. Learn about a bunch of different things.
Then, go gradually. Learn to take one small step at a time. Read Meditations by Marcus Aurelius, which has a lot of wisdom about this subject. The Tao Te Ching is another great example of literature on wisdom and patience.
If you doubt yourself and don’t want to, then you need to accumulate small victories consistently over a long period of time. Write down in one place all of the things that you have accomplished and keep adding to it all the time. Call it a brag journal if you will. Eventually the evidence will accumulate that you are capable of what you want to do.
Finally, I recommend therapy and professional coaching to really break through any underlying psychological barriers that you have e.g. pathological belief systems. My email is open if you want to talk further.
Yea I think your third paragraph about journaling accomplishments might do some good. Breaking through the mental barriers seem to be what I'm ultimately trying to accomplish.
I'll check out those books, I've heard a lot about Marcus Aurelius.
"Our model, Mip-NeRF 360, and our 'mip-NeRF 360 + iNGP' baseline were all trained on 8 NVIDIA Tesla V100-SXM2-16GB GPUs."
Even with all that, it takes almost an hour to train. A developer would need to do this for every scene, which may make it unfeasible for indies, those who would benefit the most from NERF in gaming.