Adversarial Semantic Data Augmentation for Human Pose Estimation

Abstract

Human pose estimation is the task of localizing body keypoints from still images. The state-of-the-art methods suffer from insufficient examples of challenging cases such as symmetric appearance, heavy occlusion and nearby person. To enlarge the amounts of challenging cases, previous methods augmented images by cropping and pasting image patches with weak semantics, which leads to unrealistic appearance and limited diversity. We instead propose Semantic Data Augmentation (SDA), a method that augments images by pasting segmented body parts with various semantic granularity. Furthermore, we propose Adversarial Semantic Data Augmentation (ASDA), which exploits a generative network to dynamiclly predict tailored pasting configuration. Given off-the-shelf pose estimation network as discriminator, the generator seeks the most confusing transformation to increase the loss of the discriminator while the discriminator takes the generated sample as input and learns from it. The whole pipeline is optimized in an adversarial manner. State-of-the-art results are achieved on challenging benchmarks.

References in corpus (9)

• DeepPose: Human Pose Estimation via Deep Neural Networks
• Joint Training of a Convolutional Network and a Graphical Model for Human Pose Estimation
• Human pose estimation via Convolutional Part Heatmap Regression
• Semantic Jitter: Dense Supervision for Visual Comparisons via Synthetic Images
• Rethinking on Multi-Stage Networks for Human Pose Estimation
• Devil in the Details: Towards Accurate Single and Multiple Human Parsing
• Human Pose Estimation with Spatial Contextual Information
• Deep High-Resolution Representation Learning for Human Pose Estimation
• Cascade Feature Aggregation for Human Pose Estimation

Cited by in corpus (2)

• When Human Pose Estimation Meets Robustness: Adversarial Algorithms and Benchmarks
• Data Augmentation via Structured Adversarial Perturbations