Cascade Feature Aggregation for Human Pose Estimation

TL;DR

This paper introduces a Cascade Feature Aggregation method that combines multiple hourglass networks to enhance human pose estimation accuracy by leveraging multi-stage features and fusion, demonstrating superior performance on benchmark datasets.

Contribution

The novel CFA approach cascades hourglass networks and aggregates features across stages for improved robustness and localization in pose estimation.

Findings

Outperforms state-of-the-art on MPII dataset

Achieves best results on MPII benchmark

Demonstrates robustness to occlusions and low resolution

Abstract

Human pose estimation plays an important role in many computer vision tasks and has been studied for many decades. However, due to complex appearance variations from poses, illuminations, occlusions and low resolutions, it still remains a challenging problem. Taking the advantage of high-level semantic information from deep convolutional neural networks is an effective way to improve the accuracy of human pose estimation. In this paper, we propose a novel Cascade Feature Aggregation (CFA) method, which cascades several hourglass networks for robust human pose estimation. Features from different stages are aggregated to obtain abundant contextual information, leading to robustness to poses, partial occlusions and low resolution. Moreover, results from different stages are fused to further improve the localization accuracy. The extensive experiments on MPII datasets and LIP datasets demonstrate that our proposed CFA outperforms the state-of-the-art and achieves the best performance on the state-of-the-art benchmark MPII.