Joint Coordinate Regression and Association For Multi-Person Pose Estimation, A Pure Neural Network Approach

TL;DR

This paper presents JCRA, a fast, accurate, and simple one-stage neural network for multi-person 2D pose estimation that directly predicts keypoints and associations without post-processing.

Contribution

The authors introduce a novel end-to-end network architecture with a symmetric transformer-based design that improves speed and accuracy in multi-person pose estimation.

Findings

JCRA achieves 69.2 mAP on MS COCO benchmark.

JCRA is 78% faster at inference than previous methods.

Outperforms state-of-the-art in accuracy and efficiency.

Abstract

We introduce a novel one-stage end-to-end multi-person 2D pose estimation algorithm, known as Joint Coordinate Regression and Association (JCRA), that produces human pose joints and associations without requiring any post-processing. The proposed algorithm is fast, accurate, effective, and simple. The one-stage end-to-end network architecture significantly improves the inference speed of JCRA. Meanwhile, we devised a symmetric network structure for both the encoder and decoder, which ensures high accuracy in identifying keypoints. It follows an architecture that directly outputs part positions via a transformer network, resulting in a significant improvement in performance. Extensive experiments on the MS COCO and CrowdPose benchmarks demonstrate that JCRA outperforms state-of-the-art approaches in both accuracy and efficiency. Moreover, JCRA demonstrates 69.2 mAP and is 78\% faster at inference acceleration than previous state-of-the-art bottom-up algorithms. The code for this algorithm will be publicly available.