Butterfly Effect Attack: Tiny and Seemingly Unrelated Perturbations for Object Detection
This reveals vulnerabilities in object detection systems, particularly for safety-critical applications, but is incremental as it builds on existing adversarial attack research.
The paper tackles the problem of finding tiny, seemingly unrelated image perturbations that degrade object detection performance, demonstrating that invisible perturbations on one part of an image can drastically change detection outcomes elsewhere, with transformer-based networks being more susceptible than YOLOv5.
This work aims to explore and identify tiny and seemingly unrelated perturbations of images in object detection that will lead to performance degradation. While tininess can naturally be defined using $L_p$ norms, we characterize the degree of "unrelatedness" of an object by the pixel distance between the occurred perturbation and the object. Triggering errors in prediction while satisfying two objectives can be formulated as a multi-objective optimization problem where we utilize genetic algorithms to guide the search. The result successfully demonstrates that (invisible) perturbations on the right part of the image can drastically change the outcome of object detection on the left. An extensive evaluation reaffirms our conjecture that transformer-based object detection networks are more susceptible to butterfly effects in comparison to single-stage object detection networks such as YOLOv5.