Adversarial Semantic Data Augmentation for Human Pose Estimation
This addresses data augmentation limitations for pose estimation researchers, though it is incremental as it builds on existing semantic augmentation methods.
The paper tackled the problem of insufficient challenging cases in human pose estimation by proposing Adversarial Semantic Data Augmentation (ASDA), which uses a generative network to dynamically paste segmented body parts, achieving state-of-the-art results on benchmarks.
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.