Output form of the model

elliottwu / unsup3d

(CVPR'20 Oral) Unsupervised Learning of Probably Symmetric Deformable 3D Objects from Images in the Wild

MIT License

1.19k stars 192 forks source link

Output form of the model #6

Closed VoiceBeer closed 4 years ago

VoiceBeer commented 4 years ago

Hi, thx for your work, very impressive!

I got a query about the output form of the mode. To my knowledge, the output (reconstruction) and the canonical view are all 2D images, but with depth value which could be used to reconstruct 3D volumes. Is that right? Or canonical view (reconstruction image) is a 3D volume already?

Again thank you for your work and code, looking forward to your reply :>.

elliottwu commented 4 years ago

Hi! The output of our model is depth maps in canonical viewpoint, as well as albedo, lighting, viewpoints and confidence maps. The way we render the results from various viewpoints is by constructing meshes from the depth maps, rather than 3D volumes. Maybe this demo code can be helpful: https://github.com/elliottwu/unsup3d/blob/30f4550b6bab6a520e9dd005dadac637b2fb9eb6/demo/demo.py#L196.

VoiceBeer commented 4 years ago

Hi, Many thx!

Have a good day :)