search

hassony2 / handobjectconsist

[cvpr 20] Demo, training and evaluation code for joint hand-object pose estimation in sparsely annotated videos
https://hassony2.github.io/handobjectconsist.html
MIT License
120 stars 18 forks source link

Figure7. Ho3d #12

Closed nitba closed 3 years ago

nitba commented 3 years ago

Hi @hassony2 Are the reported results in camera coordinate space or root-relative?

Best,

MengHao666 commented 3 years ago

Hi @hassony2 Are the reported results in camera coordinate space or root-relative?

Best,

Hi ,I have same confusion. So which one is right? camera coordinate space or root-relative?