search

lizhe00 / AnimatableGaussians

Code of [CVPR 2024] "Animatable Gaussians: Learning Pose-dependent Gaussian Maps for High-fidelity Human Avatar Modeling"
https://animatable-gaussians.github.io/
Other
920 stars 62 forks source link

details about training and testing on avatarrex dataset #15

Open liwenssss opened 7 months ago

liwenssss commented 7 months ago

Hi, thanks for your nice work. I have read your paper, but I did not find any information about the viewpoint settings during training and testing. I noticed that all 16 viewpoints were used during training in your code, but can you clarify which viewpoints were used during testing? For example, which viewpoints were the results in Table 2 of the paper based on? Could you provide some instructions regarding the frame and camera viewpoint configuration during testing?

lizhe00 commented 7 months ago

Details of Tab.2 are in the supplementary document. We provide "camera", "free", "front", "back", "moving" for view_setting.

liwenssss commented 7 months ago

Details of Tab.2 are in the supplementary document. We provide "camera", "free", "front", "back", "moving" for view_setting.

In the supplementary materials, you mention conducting tests from the "22010710" camera view, which also appears to be included in your training camera view?

liwenssss commented 7 months ago

image Additionally, some of the masked images contain noise. Have you applied any specific processing to address this?