论文标题
使用功能对齐可以提高对象检测中的平均精度和对抗鲁棒性
Using Feature Alignment Can Improve Clean Average Precision and Adversarial Robustness in Object Detection
论文作者
论文摘要
干净的图像中的2D对象检测是一个经过深入研究的主题，但是它针对对抗性攻击的脆弱性仍然令人担忧。现有工作通过对抗训练改善了对象探测器的鲁棒性，同时，干净图像的平均精度（AP）大大下降。在本文中，我们建议使用中间层的特征对齐可以改善对象检测的清洁AP和鲁棒性。此外，在对抗性训练的基础上，我们提供了两个特征对齐模块：知识延伸功能对齐（KDFA）模块和自我监督的功能对齐（SSFA）模块，可以指导网络生成更有效的功能。我们对Pascal VOC和MS-Coco数据集进行了广泛的实验，以验证我们提出的方法的有效性。我们的实验代码可在https://github.com/grispeut/feature-alignment.git上找到。
The 2D object detection in clean images has been a well studied topic, but its vulnerability against adversarial attack is still worrying. Existing work has improved robustness of object detectors by adversarial training, at the same time, the average precision (AP) on clean images drops significantly. In this paper, we propose that using feature alignment of intermediate layer can improve clean AP and robustness in object detection. Further, on the basis of adversarial training, we present two feature alignment modules: Knowledge-Distilled Feature Alignment (KDFA) module and Self-Supervised Feature Alignment (SSFA) module, which can guide the network to generate more effective features. We conduct extensive experiments on PASCAL VOC and MS-COCO datasets to verify the effectiveness of our proposed approach. The code of our experiments is available at https://github.com/grispeut/Feature-Alignment.git.
