How do you account for the human movement in background modeling?

MoyGcc / vid2avatar

Vid2Avatar: 3D Avatar Reconstruction from Videos in the Wild via Self-supervised Scene Decomposition (CVPR2023)

https://moygcc.github.io/vid2avatar/

MIT License

1.25k stars 104 forks source link

How do you account for the human movement in background modeling? #11

Closed xiyichen closed 1 year ago

xiyichen commented 1 year ago

I noticed that you centered the human bounding box as origin for each frame. In that case, if the human moves, wouldn't the same 3d coordinate represent different locations in the background in different frames? How is this accounted for in background modeling?

KHao123 commented 1 year ago

I have the same problem about the camera pose. How do you understand this process? Thank you!