TensoRF
awesome-NeRF
Our great sponsors
| TensoRF | awesome-NeRF | |
|---|---|---|
| 1 | 18 | |
| 1,132 | 6,238 | |
| - | 2.7% | |
| 2.9 | 6.9 | |
| 7 months ago | 21 days ago | |
| Python | TeX | |
| MIT License | MIT License |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
TensoRF
-
Nvidia Research Turns 2D Photos into 3D Scenes in the Blink of an AI
This is great, and the paper+codebase they're referring to (but not linking, here [1]) is neat too.
The research is moving fast though, so if you want something almost as fast without specialized CUDA kernels (just plain pytorch) you're in luck: https://github.com/apchenstu/TensoRF
As a bonus you also get a more compact representation of the scene.
[1] https://github.com/NVlabs/instant-ngp
awesome-NeRF
- Recommendation for a convenient NERF model to try out? And discussions...
- This is not drone footage or an iPhone video but An AI model made this. Google researchers created this 3D scene and walkthrough using just 2D images. This is called a NeRF (Anti-Aliased Grid-Based Neural Radiance Fields), where AI models can take 2D pictures and create 3D scenes.
-
New AI Tools for iPhones: Motion Capture and Environment Scanning - But What About Android Users?
And plenty others from outside Google Research. However, I wasn't aware there was a whole product making the creation of them trivial already. It's great to see honestly as I was hoping this would make the leap from research to products given how useful it is.
-
Is Nerf better than COLMAP for object reconstruction?
You can extract a mesh from a NeRF by using the marching cubes algorithm. But you'll have to texture it as well, and only after that can you be real time. NeRF training or inference isn't fast in the vanilla version. I suggest you browse https://github.com/awesome-NeRF/awesome-NeRF and look for the fast versions, this one is very fast, if you're ok with special CUDA kernels: https://nvlabs.github.io/instant-ngp/
- I volunteered to help out with the Awesome NeRF list - help me bring it up to date.
-
Nerf meshes are crap and thats normal right ?
There are implementations of NeRF that are solely based on the aim of a good export, and the work is going fast. I'm mobile right now but think this is a good place to watch, if my bookmarks are correct: https://github.com/yenchenlin/awesome-NeRF
-
Nvidia NeRF
https://github.com/yenchenlin/awesome-NeRF watch and learn from this page!
- A curated list of NeRF papers & other resources
What are some alternatives?
instant-ngp - Instant neural graphics primitives: lightning fast NeRF and more
nerf - Code release for NeRF (Neural Radiance Fields)
colmap - COLMAP - Structure-from-Motion and Multi-View Stereo
OpenSfM - Open source Structure-from-Motion pipeline
fill - Generative fill in 3D.
gaugan - Photorealistic landscape drawings using the Nvidia SPADE model
Meshroom - 3D Reconstruction Software
svox2 - Plenoxels: Radiance Fields without Neural Networks
awesome-visual-slam - :books: The list of vision-based SLAM / Visual Odometry open source, blogs, and papers
sketch-to-art - 🖼 Create artwork from your casual sketch with GAN and style transfer