mengxuyiGit

(https://github.com/pennsplatting/Next3D/blob/4756b68a5a33f15bd379288ad119a00c9c48abd4/training_avatar_texture/triplane_next3d.py#L226C9-L226C9) Shall this line be "alpha_image_side = (alpha_images[1].bool() | alpha_images[**2**].bool()).float()" in stead of "alpha_image_side = (alpha_images[1].bool() | [alpha_images[**1**].bool()).float()"] to conbine the alpha images of two sides?

warm_up training for renderer

6

Hi, as stated in paper section 3.3.1, there is a warming up for renderer. However, I cannot find this in the training code. And the update of lr across training...

How to get the quantitative results as reported in the paper

1

Great work! I am validating your pretrained model on GSO dataset, but I cannot get as high PSNR as reported in your paper due to the camera distance & object...

RuntimeError: CUDA error: an illegal memory access was encountered

9

I wil l encounter **illegal memory access** for all **datatsets** and at **different places**: for MipNerf360, it trained for 10 iters and report error at "visibility_filter" : radii > 0,...

About rendered normal

1

Is the rendered normal representing the blended normal at **alpha=0.5** as the depth surface, or it accumulates to **alpha=1.0** which is the full density

Hi authors, thanks for this great work! I am trying to **download the data**, but it seems expired and lead to the below webpage:  Could you please help with...