Jointly Optimize Data Augmentation and Network Training: Adversarial Data Augmentation in Human Pose Estimation
This addresses the limitation of treating data augmentation and network training as isolated processes, offering a novel approach for researchers and practitioners in computer vision, particularly for human pose estimation, though it is incremental as it builds on existing adversarial and augmentation techniques.
The paper tackles the problem of overfitting in deep neural networks by jointly optimizing data augmentation and network training, introducing adversarial data augmentation to generate 'hard' augmentations online, which significantly improves state-of-the-art models in human pose estimation without extra data.
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.