EfficientPose: Efficient Human Pose Estimation with Neural Architecture Search

TL;DR

EfficientPose introduces a neural architecture search-based framework for human pose estimation that optimizes efficiency and accuracy, enabling real-time performance on resource-limited devices.

Contribution

The paper presents a novel neural architecture search approach to design efficient backbones and heads specifically for human pose estimation, reducing computation without significant accuracy loss.

Findings

Small model achieves 88.1% PCKh@0.5 on MPII with 0.65 GFLOPs.

Large model reaches competitive accuracy with only 2 GFLOPs.

Outperforms many existing models in efficiency and accuracy trade-offs.

Abstract

Human pose estimation from image and video is a vital task in many multimedia applications. Previous methods achieve great performance but rarely take efficiency into consideration, which makes it difficult to implement the networks on resource-constrained devices. Nowadays real-time multimedia applications call for more efficient models for better interactions. Moreover, most deep neural networks for pose estimation directly reuse the networks designed for image classification as the backbone, which are not yet optimized for the pose estimation task. In this paper, we propose an efficient framework targeted at human pose estimation including two parts, the efficient backbone and the efficient head. By implementing the differentiable neural architecture search method, we customize the backbone network design for pose estimation and reduce the computation cost with negligible accuracy degradation. For the efficient head, we slim the transposed convolutions and propose a spatial information correction module to promote the performance of the final prediction. In experiments, we evaluate our networks on the MPII and COCO datasets. Our smallest model has only 0.65 GFLOPs with 88.1% PCKh@0.5 on MPII and our large model has only 2 GFLOPs while its accuracy is competitive with the state-of-the-art large model, i.e., HRNet with 9.5 GFLOPs.