EvoPose2D: Pushing the Boundaries of 2D Human Pose Estimation using Neuroevolution

Neural architecture search has proven to be highly effective in the design of computationally efficient, task-specific convolutional neural networks across several areas of computer vision. In 2D human pose estimation, however, its application has been limited by high computational demands... (read more)

PDF Abstract

Datasets


Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK BENCHMARK
Keypoint Detection COCO EvoPose2D-L Validation AP 77.5 # 1
Test AP 76.8 # 1
Multi-Person Pose Estimation COCO EvoPose2D-L Validation AP 77.5 # 2
Test AP 76.8 # 1
Pose Estimation COCO test-dev EvoPose2D-L AP 76.8 # 3
AP50 92.5 # 5
AP75 84.3 # 3
APL 82.5 # 3
APM 73.5 # 3
AR 81.7 # 3

Methods used in the Paper


METHOD TYPE
🤖 No Methods Found Help the community by adding them if they're not listed; e.g. Deep Residual Learning for Image Recognition uses ResNet