search

zhangboshen / A2J

Code for paper "A2J: Anchor-to-Joint Regression Network for 3D Articulated Pose Estimation from a Single Depth Image". ICCV2019
MIT License
289 stars 46 forks source link

Data augmentation problem #26

Closed cherryjm closed 4 years ago

cherryjm commented 4 years ago

Hi, @zhangboshen ! Thank you for this work! When I go through scr_train/nyu.py, I notice there is a parameter called RandomOffsetDepth. https://github.com/zhangboshen/A2J/blob/5848571cbb9d1911fc416ae6db186d9a8b0b5b8e/src_train/nyu.py#L153 It seems that all values in it are 0. And it has never been used after this line. Is it necessary to apply this augmentation?

zhangboshen commented 4 years ago

Hi, @zhangboshen ! Thank you for this work! When I go through scr_train/nyu.py, I notice there is a parameter called RandomOffsetDepth. https://github.com/zhangboshen/A2J/blob/5848571cbb9d1911fc416ae6db186d9a8b0b5b8e/src_train/nyu.py#L153

It seems that all values in it are 0. And it has never been used after this line. Is it necessary to apply this augmentation?

I am sorry I can't recall why this line of code exists. And since this line is not used anywhere, you can just comment it.