Learning 3D Human Pose from Structure and Motion


Video Video
Pose Pose



People


Rishabh
Dabral
Anurag
Mundhada
Uday
Kusupati
Safeer
Afaque
Abhishek
Sharma
Arjun
Jain

Abstract

3D human pose estimation from a single image is a challenging problem, especially for in-the-wild settings due to the lack of 3D annotated data. We propose two anatomically inspired loss functions and use them with a weakly-supervised learning framework to jointly learn from large-scale in-the-wild 2D and indoor/synthetic 3D data. We also present a simple temporal network that exploits temporal and structural cues present in predicted pose sequences to temporally harmonize the pose estimations. We carefully analyze the proposed contributions through loss surface visualizations and sensitivity analysis to facilitate deeper understanding of their working mechanism. Jointly, the two networks capture the anatomical constraints in static and kinetic states of the human body. Our complete pipeline improves the state-of-the-art by 11.8% and 12% on Human3.6M and MPI-INF-3DHP, respectively, and runs at 30 FPS on a commodity graphics card.

Materials for Download


 Paper (arXiv link)
Supplementary Material

Visual Results



Citation

Rishabh Dabral, Anurag Mundhada, Uday Kusupati, Safeer Afaque, Abhishek Sharma, Arjun Jain
Learning 3D Human Pose from Structure and Motion
European Conference on Computer Vision 2018 


					@article{Dabral:ECCV:2018,
					title  	  = {Learning 3D Human Pose from Structure and Motion}
					author    = {Dabral, Rishabh and 
						     Mundhada, Anurag and
						     Kusupati, Uday and 
						     Afaque, Safeer and
					             Sharma, Abhishek and
						     Jain, Arjun},
					booktitle = {Computer Vision -- ECCV 2018},
					series    = {Lecture Notes in Computer Science},
					publisher = {Springer International Publishing},
					year	  = {2018}
					}