OmniPose: A Multi-Scale Framework for Multi-Person Pose Estimation

18 Mar 2021  ·  Bruno Artacho, Andreas Savakis ·

We propose OmniPose, a single-pass, end-to-end trainable framework, that achieves state-of-the-art results for multi-person pose estimation. Using a novel waterfall module, the OmniPose architecture leverages multi-scale feature representations that increase the effectiveness of backbone feature extractors, without the need for post-processing. OmniPose incorporates contextual information across scales and joint localization with Gaussian heatmap modulation at the multi-scale feature extractor to estimate human pose with state-of-the-art accuracy. The multi-scale representations, obtained by the improved waterfall module in OmniPose, leverage the efficiency of progressive filtering in the cascade architecture, while maintaining multi-scale fields-of-view comparable to spatial pyramid configurations. Our results on multiple datasets demonstrate that OmniPose, with an improved HRNet backbone and waterfall module, is a robust and efficient architecture for multi-person pose estimation that achieves state-of-the-art results.

PDF Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Pose Estimation COCO OmniPose (WASPv2) AP 79.5 # 1
AR 81.9 # 2
AP50 93.6 # 1
AP75 85.9 # 1
APM 76 # 1
APL 84.6 # 1
Pose Estimation COCO test-dev OmniPose (WASPv2) AP 76.4 # 16
AP50 92.6 # 13
AP75 83.7 # 16
APL 82.6 # 8
APM 72.6 # 13
AR 81.2 # 16
Pose Estimation Leeds Sports Poses OmniPose PCK 99.5% # 1
Pose Estimation MPII OmniPose (WASPv2) PCKh@0.2 92.3 # 1
Pose Estimation UPenn Action OmniPose Mean PCK@0.2 99.4 # 1