Texture- and Shape-based Adversarial Attacks for Overhead Image Vehicle Detection
This work addresses the reliability of object detection models in overhead imagery for applications like surveillance or mapping, but it is incremental as it builds on existing adversarial attack methods with added practical constraints.
The paper tackled the problem of adversarial attacks on vehicle detection in aerial images by proposing realistic constraints on texture and shape modifications, finding that more practical modifications tend to be less effective, with experiments showing trade-offs in performance across three detector architectures.
Detecting vehicles in aerial images is difficult due to complex backgrounds, small object sizes, shadows, and occlusions. Although recent deep learning advancements have improved object detection, these models remain susceptible to adversarial attacks (AAs), challenging their reliability. Traditional AA strategies often ignore practical implementation constraints. Our work proposes realistic and practical constraints on texture (lowering resolution, limiting modified areas, and color ranges) and analyzes the impact of shape modifications on attack performance. We conducted extensive experiments with three object detector architectures, demonstrating the performance-practicality trade-off: more practical modifications tend to be less effective, and vice versa. We release both code and data to support reproducibility at https://github.com/humansensinglab/texture-shape-adversarial-attacks.