Chen Wang

How to set the near and far plane in NeRF network?

1

Thanks for your excellent work. I am curious why you set the ray_near and ray_end to 0.88 and 1.12? (and for other variables like h_stddev etc.) Is that set empirically?

For example, in scene 8WUmhLawc2A, the undistorted pose is ``scan 01b439d39a8f412fa1837be7afb45254_d0_0.png 01b439d39a8f412fa1837be7afb45254_i0_0.jpg 0.863804 0.361417 0.351025 -5.69715 0.502959 -0.577708 -0.642872 2.5578 -0.0295543 0.731867 -0.680806 1.41593 0 0 0 1``, but in...

Thanks for the excellent work. I noticed that you trained other baselines (e.g. PixelNeRF) on the same dataset as yours, but I found that adapting the extra dataset on pixelnerf...

Hi, I see your code includes the dataset loader for LLFF dataset. Could you share any tips on how to run your code on LLFF dataset?

Error in adop_train

2

after throwing an instance of std::runtime_error what(): nvrtc: error: failed to open libnvrtc-builtins.so.11.2. Make sure that libnvrtc-builtins.so.11.2 is installed correctly. nvrtc compilation failed After trying setting the LD_LIBRARY_PATH as suggested,...

Why latent encode in stablediffusion needs grad?

2

I think the encoding part is part of the diffusion model and doesn't need to train. But why you are training here?

Unmatched performance with README

3

I trained with the command `python main.py --text 'a squirrel' --workspace trial -O --eval_interval 10`. However, my results turned out to be much worse than the squirrel shown in the...

Weird results when testing on customized images

3

Thanks for the amazing work! I tried to run the gaussian reconstruction part on 4 images I rendered, the four images have elevation and azimuth of: (30, 30), (-20, 90),...

Questions on calculating the distance_mean (depth)

1

Thanks for the excellent work of mipnerf360. I am a little confused about the depth loss in mipnerf360. The current way of calculating depth is: ` rendering['distance_mean'] = ( jnp.clip(...