100DOH

Abstract

Hands are the central means by which humans manipulate their world and being able to reliably extract hand state information from Internet videos of humans engaged in their hands has the potential to pave the way to systems that can learn from petabytes of video data.

This paper proposes steps towards this by inferring a rich representation of hands engaged in interaction method that includes: hand location, side, contact state, and a box around the object in contact. To support this effort, we gather a large-scale dataset of hands in contact with objects consisting of 131 days of footage as well as a 100K annotated hand-contact video frame dataset. The learned model on this dataset can serve as a foundation for hand-contact understanding in videos. We quantitatively evaluate it both on its own and in service of predicting and learning from 3D meshes of human hands.

Paper

[Paper]

[Supplemental]

[Code]

[Slides]

[Talk]

Citation


@INPROCEEDINGS{Shan20, 
    author = {Shan, Dandan and Geng, Jiaqi and Shu, Michelle  and Fouhey, David},
    title = {Understanding Human Hands in Contact at Internet Scale},
    booktitle = CVPR, 
    year = {2020} 
}

Results

Acknowledgement

This work was supported by: the Advanced Machine Learning Collaborative Grant from Procter & Gamble in collaboration with Matthew Barker, PhD; and a gift from Nokia Solutions and Networks Oy.

DS thanks Mohamed El Banani, Karan Desai, Richard Higgins, Linyi Jin, Nilesh Kulkarni, Shengyi Qian, Chris Rockwell for advice throughout the process and all the feedbacks from our users ❤️❤️❤️.