Robust and Accurate Object Detection via Adversarial Learning

TL;DR

This paper introduces an adversarial augmentation method for object detection that enhances accuracy and robustness by dynamically selecting adversarial examples during fine-tuning, outperforming existing augmentation techniques.

Contribution

The work proposes a novel adversarial augmentation approach that improves object detector performance and robustness, outperforming model-agnostic methods like AutoAugment.

Findings

Boosts EfficientDet mAP by +1.1 on COCO

Enhances robustness against natural distortions by +3.8 mAP

Improves domain shift robustness by +1.3 mAP

Abstract

Data augmentation has become a de facto component for training high-performance deep image classifiers, but its potential is under-explored for object detection. Noting that most state-of-the-art object detectors benefit from fine-tuning a pre-trained classifier, we first study how the classifiers' gains from various data augmentations transfer to object detection. The results are discouraging; the gains diminish after fine-tuning in terms of either accuracy or robustness. This work instead augments the fine-tuning stage for object detectors by exploring adversarial examples, which can be viewed as a model-dependent data augmentation. Our method dynamically selects the stronger adversarial images sourced from a detector's classification and localization branches and evolves with the detector to ensure the augmentation policy stays current and relevant. This model-dependent augmentation generalizes to different object detectors better than AutoAugment, a model-agnostic augmentation policy searched based on one particular detector. Our approach boosts the performance of state-of-the-art EfficientDets by +1.1 mAP on the COCO object detection benchmark. It also improves the detectors' robustness against natural distortions by +3.8 mAP and against domain shift by +1.3 mAP. Models are available at https://github.com/google/automl/tree/master/efficientdet/Det-AdvProp.md