Relightable3DGaussian
Relightable3DGaussian copied to clipboard
NJU-3DV
[ECCV2024] Relightable 3D Gaussian: Real-time Point Cloud Relighting with BRDF Decomposition and Ray Tracing
``` trace_results = raytracer.trace_visibility( rays_o, rays_d, means3D, cov_inv, opacity, normal) visibility = trace_results['visibility'] ``` say, N: number of gaussians, S: number of sampled rays From my understanding, - rays_o is...
Hi Thanks for your excellent work. Could you please provide some code for ref-nerf synthetic dataset and relighting on them? I've tried it by myself but I failed due to...
First of all, I really want to thank you for your work. It's a great job! The problem I ran into was the time it took me to train stage2...
I would like to use a different dataset from the following link: [Gaussian Grouping - Figurines](https://huggingface.co/mqye/Gaussian-Grouping/blob/main/data/lerf_mask/figurines.zip). However, I encountered an issue and would appreciate some help. The structure of the...
Thanks for sharing the codes. I have a question that since the pytorch was installed with cudatoolkit==11.6, in the next step of bvh installation, do we need to change the...
Hello, thank you for your great work! I was trying to reproduce the 'Composition and Relighting' part in the README, but I found that the ground.ply download link is not...
Firstly, I'd like to express my appreciation for the outstanding work you've done on this project. It's truly impressive. I'm curious about the files light_transform.json, trajectory.json, and transform.json. Are these...
Firstly, I would like to express my sincere appreciation for the outstanding work you have contributed to the community. Your project is truly inspiring. As a beginner in the field,...
Modified `gui.py` by adding a drop-down for changing environment map lighting. This allows us to visualize the relighting results on-the-fly in real-time. While `global_light` is correctly changed, there is no...
Hello, excellent work! I have a question regarding your paper. Why bind `normal` and `color` to the same Gaussian splats and optimize both parameters concurrently? What advantages does this approach...
