Swin-Pose: Swin Transformer Based Human Pose Estimation

TL;DR

This paper introduces Swin-Pose, a transformer-based human pose estimation model that leverages a Swin Transformer backbone and feature pyramid fusion to outperform CNN-based methods.

Contribution

The paper presents a novel transformer-based architecture with feature pyramid fusion for improved human pose estimation accuracy.

Findings

Achieves better performance than state-of-the-art CNN models

Utilizes pre-trained Swin Transformer as backbone

Demonstrates effectiveness of transformer architecture in pose estimation

Abstract

Convolutional neural networks (CNNs) have been widely utilized in many computer vision tasks. However, CNNs have a fixed reception field and lack the ability of long-range perception, which is crucial to human pose estimation. Due to its capability to capture long-range dependencies between pixels, transformer architecture has been adopted to computer vision applications recently and is proven to be a highly effective architecture. We are interested in exploring its capability in human pose estimation, and thus propose a novel model based on transformer architecture, enhanced with a feature pyramid fusion structure. More specifically, we use pre-trained Swin Transformer as our backbone and extract features from input images, we leverage a feature pyramid structure to extract feature maps from different stages. By fusing the features together, our model predicts the keypoint heatmap. The experiment results of our study have demonstrated that the proposed transformer-based model can achieve better performance compared to the state-of-the-art CNN-based models.