Jointly Optimize Data Augmentation and Network Training: Adversarial Data Augmentation in Human Pose Estimation

TL;DR

This paper introduces adversarial data augmentation that jointly optimizes augmentation and network training, leading to improved human pose estimation performance without extra data.

Contribution

It proposes a novel adversarial augmentation framework that dynamically generates challenging augmentations during training, enhancing model robustness and accuracy.

Findings

Significant performance improvements on human pose estimation benchmarks.

Effective joint optimization of augmentation and training processes.

No additional data required for improved results.

Abstract

Random data augmentation is a critical technique to avoid overfitting in training deep neural network models. However, data augmentation and network training are usually treated as two isolated processes, limiting the effectiveness of network training. Why not jointly optimize the two? We propose adversarial data augmentation to address this limitation. The main idea is to design an augmentation network (generator) that competes against a target network (discriminator) by generating `hard' augmentation operations online. The augmentation network explores the weaknesses of the target network, while the latter learns from `hard' augmentations to achieve better performance. We also design a reward/penalty strategy for effective joint training. We demonstrate our approach on the problem of human pose estimation and carry out a comprehensive experimental analysis, showing that our method can significantly improve state-of-the-art models without additional data efforts.