Official PyTorch implementation of "I2L-MeshNet: Image-to-Lixel Prediction Network for Accurate 3D Human Pose and Mesh Estimation from a Single RGB Image", ECCV 2020
MIT License
708
stars
130
forks
source link
About the absolute 3D coordinates recovery of the hand #106
Hi, I have a question about the absolute 3D coordinates recovery of the hand.
If I want to get the absolute 3D coordinates of the hand, should I use the RootNet to predict the root depth, while MeshNet predicts the root-relative 3D coordinates. Or is there some other methods to get the absolute 3D coordinates?
Looking forward to getting any kind of advice.
mks0601
commented
2 years ago
Hi, I have a question about the absolute 3D coordinates recovery of the hand. If I want to get the absolute 3D coordinates of the hand, should I use the RootNet to predict the root depth, while MeshNet predicts the root-relative 3D coordinates. Or is there some other methods to get the absolute 3D coordinates? Looking forward to getting any kind of advice.