arXiv Analytics

Sign in

arXiv:1710.10192 [cs.CV]AbstractReferencesReviewsResources

Dual Path Networks for Multi-Person Human Pose Estimation

Guanghan Ning, Zhihai He

Published 2017-10-27Version 1

The task of multi-person human pose estimation in natural scenes is quite challenging. Existing methods include both top-down and bottom-up approaches. The main advantage of bottom-up methods is its excellent tradeoff between estimation accuracy and computational cost. We follow this path and aim to design smaller, faster, and more accurate neural networks for the regression of keypoints and limb association vectors. These two regression tasks are naturally dependent on each other. In this work, we propose a dual-path network specially designed for multi-person human pose estimation, and compare our performance with the openpose network in aspects of model size, forward speed, and estimation accuracy.

Comments: ICCV 2017 Workshop on PoseTrack Challenge. Challenge results available at: https://posetrack.net/workshops/iccv2017/posetrack-challenge-results.html
Categories: cs.CV
Related articles: Most relevant | Search more
arXiv:2310.01259 [cs.CV] (Published 2023-10-02, updated 2023-10-03)
Faster and Accurate Neural Networks with Semantic Inference
arXiv:1307.6962 [cs.CV] (Published 2013-07-26, updated 2013-12-01)
Reduced egomotion estimation drift using omnidirectional views
arXiv:1810.08338 [cs.CV] (Published 2018-10-19)
Multi-Domain Pose Network for Multi-Person Pose Estimation and Tracking