EvoPose2D: Pushing the Boundaries of 2D Human Pose Estimation using Accelerated Neuroevolution with Weight Transfer

17 Nov 2020  ยท  William McNally, Kanav Vats, Alexander Wong, John McPhee ยท

Neural architecture search has proven to be highly effective in the design of efficient convolutional neural networks that are better suited for mobile deployment than hand-designed networks. Hypothesizing that neural architecture search holds great potential for human pose estimation, we explore the application of neuroevolution, a form of neural architecture search inspired by biological evolution, in the design of 2D human pose networks for the first time. Additionally, we propose a new weight transfer scheme that enables us to accelerate neuroevolution in a flexible manner. Our method produces network designs that are more efficient and more accurate than state-of-the-art hand-designed networks. In fact, the generated networks process images at higher resolutions using less computation than previous hand-designed networks at lower resolutions, allowing us to push the boundaries of 2D human pose estimation. Our base network designed via neuroevolution, which we refer to as EvoPose2D-S, achieves comparable accuracy to SimpleBaseline while being 50% faster and 12.7x smaller in terms of file size. Our largest network, EvoPose2D-L, achieves new state-of-the-art accuracy on the Microsoft COCO Keypoints benchmark, is 4.3x smaller than its nearest competitor, and has similar inference speed. The code is publicly available at https://github.com/wmcnally/evopose2d.

PDF Abstract

Results from the Paper


 Ranked #1 on Multi-Person Pose Estimation on MS COCO (Validation AP metric)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Pose Estimation COCO test-dev EvoPose2D-L AP 76.8 # 12
AP50 92.5 # 15
AP75 84.3 # 11
APL 82.5 # 9
APM 73.5 # 11
AR 81.7 # 10
Multi-Person Pose Estimation MS COCO EvoPose2D-L Validation AP 77.5 # 1
Test AP 76.8 # 1
Keypoint Detection MS COCO EvoPose2D-L(512x384) Validation AP 77.5 # 3
Test AP 76.8 # 3

Methods


No methods listed for this paper. Add relevant methods here