facebookresearch / ava-256

Train universal codec avatars
Other
96 stars 4 forks source link

About getting depth information from the provided data #17

Open longyangqi opened 1 month ago

longyangqi commented 1 month ago

Awesome work ! I wonder if we can get the depth map of the faces from the provided data?

Many Thanks!

una-dinosauria commented 2 weeks ago

We don't have gt depth for this data. You can use an off-the-shelf depth estimator like Sapiens, or use the provided meshes as a proxy.

longyangqi commented 2 weeks ago

We don't have gt depth for this data. You can use an off-the-shelf depth estimator like Sapiens, or use the provided meshes as a proxy.

Thanks for your reply. I have tried to use the mesh as stated in https://github.com/facebookresearch/ava-256/issues/22 Another question, how do you get the meshes? I also want to collect some data in my environment.