Greit-HRNet: Grouped Lightweight High-Resolution Network for Human Pose Estimation

TL;DR

Greit-HRNet is a novel lightweight high-resolution network for human pose estimation that uses grouped modules and global spatial weighting to improve efficiency and accuracy.

Contribution

The paper introduces Greit-HRNet with grouped channel and spatial weighting modules, enhancing global information capture and stability in lightweight networks.

Findings

Outperforms state-of-the-art lightweight networks on MS-COCO and MPII datasets.

Effectively maintains high-resolution features with increased network depth.

Improves global spatial information extraction and stability.

Abstract

As multi-scale features are necessary for human pose estimation tasks, high-resolution networks are widely applied. To improve efficiency, lightweight modules are proposed to replace costly point-wise convolutions in high-resolution networks, including channel weighting and spatial weighting methods. However, they fail to maintain the consistency of weights and capture global spatial information. To address these problems, we present a Grouped lightweight High-Resolution Network (Greit-HRNet), in which we propose a Greit block including a group method Grouped Channel Weighting (GCW) and a spatial weighting method Global Spatial Weighting (GSW). GCW modules group conditional channel weighting to make weights stable and maintain the high-resolution features with the deepening of the network, while GSW modules effectively extract global spatial information and exchange information across channels. In addition, we apply the Large Kernel Attention (LKA) method to improve the whole efficiency of our Greit-HRNet. Our experiments on both MS-COCO and MPII human pose estimation datasets demonstrate the superior performance of our Greit-HRNet, outperforming other state-of-the-art lightweight networks.