Common Corruptions for Enhancing and Evaluating Robustness in Air-to-Air Visual Object Detection

TL;DR

This paper introduces a new robustness benchmark dataset for air-to-air object detection, evaluates various detectors under corruptions, and shows that fine-tuning improves real-world performance.

Contribution

It creates the first robustness benchmark dataset for aerial object detection, evaluates multiple detectors under corruptions, and demonstrates benefits of fine-tuning on synthetic data.

Findings

YOLO detectors are more robust to corruptions.

Transformer-based detectors are highly vulnerable.

Fine-tuning on synthetic data improves real-world detection.

Abstract

The main barrier to achieving fully autonomous flights lies in autonomous aircraft navigation. Managing non-cooperative traffic presents the most important challenge in this problem. The most efficient strategy for handling non-cooperative traffic is based on monocular video processing through deep learning models. This study contributes to the vision-based deep learning aircraft detection and tracking literature by investigating the impact of data corruption arising from environmental and hardware conditions on the effectiveness of these methods. More specifically, we designed $7$ types of common corruptions for camera inputs taking into account real-world flight conditions. By applying these corruptions to the Airborne Object Tracking (AOT) dataset we constructed the first robustness benchmark dataset named AOT-C for air-to-air aerial object detection. The corruptions included in this dataset cover a wide range of challenging conditions such as adverse weather and sensor noise. The second main contribution of this letter is to present an extensive experimental evaluation involving $8$ diverse object detectors to explore the degradation in the performance under escalating levels of corruptions (domain shifts). Based on the evaluation results, the key observations that emerge are the following: 1) One-stage detectors of the YOLO family demonstrate better robustness, 2) Transformer-based and multi-stage detectors like Faster R-CNN are extremely vulnerable to corruptions, 3) Robustness against corruptions is related to the generalization ability of models. The third main contribution is to present that finetuning on our augmented synthetic data results in improvements in the generalisation ability of the object detector in real-world flight experiments.