search

namepllet / HandOccNet

Offical pytorch implementation of "HandOccNet: Occlusion-Robust 3D Hand Mesh Estimation Network", CVPR 2022.
158 stars 31 forks source link

Question about the result on FPHA dataset, is that after procrustes alignment? #23

Open LiuYi-Up opened 1 year ago

LiuYi-Up commented 1 year ago

Hi, thanks for your excellent work.

I found the sentence "All metrics are obtained after the procrustes alignment. " on your paper, but I want to make sure is the result on FPHA also evaluated after procrustes alignment?

Look forward your answer. Best wishes.

karta2155802 commented 1 year ago

Hi, I found that the official FPHA dataset only provide joints 3d location for hand pose annotations. I wonder how did you train on HandOccNet?

LiuYi-Up commented 1 year ago

Hi, I found that the official FPHA dataset only provide joints 3d location for hand pose annotations. I wonder how did you train on HandOccNet?

I use the points GT to supervise the model only.

karta2155802 commented 1 year ago

Hi, I found that the official FPHA dataset only provide joints 3d location for hand pose annotations. I wonder how did you train on HandOccNet?

I use the points GT to supervise the model only.

Thanks for the quick reply. Did you get a similar result with the paper?

LiuYi-Up commented 1 year ago

Hi, I found that the official FPHA dataset only provide joints 3d location for hand pose annotations. I wonder how did you train on HandOccNet?

I use the points GT to supervise the model only.

Thanks for the quick reply. Did you get a similar result with the paper?

little better