ADA: Adversarial Data Augmentation for Object Detection
ADA: Adversarial Data Augmentation for Object Detection
复制标题
DOI：
10.1109/wacv.2019.00137
复制
发表时间：
2019-01
期刊：
影响因子：
--
通讯作者：
Sima Behpour;Kris M. Kitani;Brian D. Ziebart
中科院分区：
文献类型：
--
作者：
Sima Behpour;Kris M. Kitani;Brian D. Ziebart
使用地面真实数据的随机扰动，例如边界框的随机平移或缩放，是用于数据增强的常见启发式方法，已被证明可以防止过度拟合和改进泛化。由于数据增强的设计在很大程度上是由已报告的最佳实践指导的，因此很难理解这些设计选择是否为最佳。为了提供一个更有原则的视角，我们发展了一个博弈论的解释，在对象检测的背景下数据增加。我们的目标是找到一种最优的地面真实数据的对抗性扰动(即，最坏情况的扰动)，迫使对象边界盒预测器从扰动示例的最难分布中学习，以获得更好的测试时间性能。我们证明了博弈论解(纳什均衡)既提供了最优预报器，又提供了最优数据增长分布。我们表明，我们的对抗性训练预测器的方法可以显著提高目标检测任务的测试时间性能。在ImageNet、Pascal VOC和MS-Coco目标检测任务上，与性能最好的数据增强方法相比，我们的对抗性方法分别提高了16%、5%和2%的性能。
The use of random perturbations of ground truth data, such as random translation or scaling of bounding boxes, is a common heuristic used for data augmentation that has been shown to prevent overfitting and improve generalization. Since the design of data augmentation is largely guided by reported best practices, it is difficult to understand if those design choices are optimal. To provide a more principled perspective, we develop a game-theoretic interpretation of data augmentation in the context of object detection. We aim to find an optimal adversarial perturbations of the ground truth data (i.e., the worst case perturbations) that forces the object bounding box predictor to learn from the hardest distribution of perturbed examples for better test-time performance. We establish that the game-theoretic solution (Nash equilibrium) provides both an optimal predictor and optimal data augmentation distribution. We show that our adversarial method of training a predictor can significantly improve test-time performance for the task of object detection. On the ImageNet, Pascal VOC and MS-COCO object detection tasks, our adversarial approach improves performance by about 16%, 5%, and 2% respectively compared to the best performing data augmentation methods.