Simple and Lightweight Human Pose Estimation

TL;DR

This paper introduces a lightweight human pose estimation network that achieves competitive accuracy with significantly reduced model size and computational cost, making it suitable for practical deployment on resource-constrained devices.

Contribution

The paper proposes a novel lightweight bottleneck block and a pose network architecture that drastically reduces model size and FLOPs while maintaining high accuracy.

Findings

LPN-50 achieves 68.7 AP on COCO test-dev.

Model size is only 9% of ResNet50-based methods.

Inference speed reaches 17 FPS on CPU.

Abstract

Recent research on human pose estimation has achieved significant improvement. However, most existing methods tend to pursue higher scores using complex architecture or computationally expensive models on benchmark datasets, ignoring the deployment costs in practice. In this paper, we investigate the problem of simple and lightweight human pose estimation. We first redesign a lightweight bottleneck block with two non-novel concepts: depthwise convolution and attention mechanism. And then, based on the lightweight block, we present a Lightweight Pose Network (LPN) following the architecture design principles of SimpleBaseline. The model size (#Params) of our small network LPN-50 is only 9% of SimpleBaseline(ResNet50), and the computational complexity (FLOPs) is only 11%. To give full play to the potential of our LPN and get more accurate predicted results, we also propose an iterative training strategy and a model-agnostic post-processing function Beta-Soft-Argmax. We empirically demonstrate the effectiveness and efficiency of our methods on the benchmark dataset: the COCO keypoint detection dataset. Besides, we show the speed superiority of our lightweight network at inference time on a non-GPU platform. Specifically, our LPN-50 can achieve 68.7 in AP score on the COCO test-dev set, with only 2.7M parameters and 1.0 GFLOPs, while the inference speed is 17 FPS on an Intel i7-8700K CPU machine.