namepllet / HandOccNet

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

CAN NOT be applied to real-world situations #38

Open TekamanBlade opened 1 week ago

TekamanBlade commented 1 week ago

For this method to achieve accurate hand pose estimation under occlusion conditions, it first requires an accurate bounding box to cover the complete hand area. However, the problem of obtaining the hand area under occlusion conditions itself demands the ability to accurately estimate the hand pose under occlusion. Therefore, this method simply cannot be applied to real-world scenarios. Am I right?