Domes to drones: Self-supervised active triangulation for 3d human pose reconstruction supplementary material

Research output: Contribution to journalArticle

Abstract

This supplementary provides more insight into our ACTOR model and experimental setup. Section §1 describes the details of the network architecture, implementation, and hyperparameters. §2 elaborates on how we match 2d pose estimates in space and time using instance features. In §3 we provide 2d reprojection errors onto 2d OpenPose [2] estimates on the Panoptic test splits. Finally, §4 describes further dataset details.

Details

Authors
Organisations
External organisations
  • Google Inc.
Research areas and keywords

Subject classification (UKÄ) – MANDATORY

  • Mathematics
Original languageEnglish
JournalAdvances in Neural Information Processing Systems
Volume32
Publication statusPublished - 2019 Jan 1
Publication categoryResearch
Peer-reviewedYes
Event33rd Annual Conference on Neural Information Processing Systems, NeurIPS 2019 - Vancouver, Canada
Duration: 2019 Dec 82019 Dec 14