Part-aware measurement for robust multi-view multi-human 3d pose estimation and tracking
Journal
IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
Pages
1472-1481
Date Issued
2021
Author(s)
Abstract
This paper introduces an approach for multi-human 3D pose estimation and tracking based on calibrated multi-view. The main challenge lies in finding the cross-view and temporal correspondences correctly even when several human pose estimations are noisy. Compare to previous solutions that construct 3D poses from multiple views, our approach takes advantage of temporal consistency to match the 2D poses estimated with previously constructed 3D skeletons in every view. Therefore cross-view and temporal associations are accomplished simultaneously. Since the performance suffers from mistaken association and noisy predictions, we design two strategies for aiming better correspondences and 3D reconstruction. Specifically, we propose a part-aware measurement for 2D-3D association and a filter that can cope with 2D outliers during reconstruction. Our approach is efficient and effective comparing to state-of-the-art methods; it achieves competitive results on two benchmarks: 96.8% on Campus and 97.4% on Shelf. Moreover, we extends the length of Campus evaluation frames to be more challenging and our proposal also reach well-performed result. The code will be available at https://git.io/JO4KE. ? 2021 IEEE.
Subjects
Computer vision
Gesture recognition
Petroleum reservoir evaluation
3D pose estimation
3D pose tracking
3D skeleton
Estimation and tracking
Human pose estimations
Multi-views
Multiple views
Performance
Temporal association
Temporal consistency
Image reconstruction
Type
conference paper