What was the fitting procedure for MANO parameters?

facebookresearch / InterHand2.6M

Official PyTorch implementation of "InterHand2.6M: A Dataset and Baseline for 3D Interacting Hand Pose Estimation from a Single RGB Image", ECCV 2020

Other

687 stars 91 forks source link

What was the fitting procedure for MANO parameters? #20

Closed pablovela5620 closed 3 years ago

pablovela5620 commented 3 years ago

Would love to get a bit more info as to how you fit the MANO hand model to the hand. Was it a similar process to what was done in the FreiHand dataset (multiview loss function for 2d/3d/segmentation) or something different like the SPIN algorithm for full-body mesh estimation? I didn't see anything about a fitting procedure in the original paper. Thanks!

mks0601 commented 3 years ago

That is a part of my ongoing CVPR work. I'll upload it on arxiv maybe right after cvpr submission (mid of Nov.).

pablovela5620 commented 3 years ago

Understood, appreciate the info! Looking forward to seeing it

mks0601 commented 3 years ago

Hi, my new work used to get the MANO parameters is available on arxiv

pablovela5620 commented 3 years ago

Congrats on the submissions! I'd love to ask you some questions on the paper, but I don't know if this is the best place (or if I should wait for the code release on the neural annotator) please let me know. Again congrats, looks like this will solve alot of the current dataset generation problems!

mks0601 commented 3 years ago

the codes will be available once it is accepted, so maybe you can ask questions by email?