A Unified Spatial Alignment Framework for Highly Transferable Transformation-Based Attacks on Spatially Structured Tasks
This addresses a critical limitation in adversarial attacks for computer vision tasks, enabling more effective attacks on structured tasks, though it is incremental as it builds on existing transformation-based methods.
The paper tackles the problem of transformation-based adversarial attacks performing poorly on spatially structured tasks like semantic segmentation and object detection by proposing a unified Spatial Alignment Framework (SAF) that synchronously transforms labels with inputs, resulting in significant performance degradation, such as reducing average mIoU on Cityscapes from 24.50 to 11.34 and average mAP on COCO from 17.89 to 5.25.
Transformation-based adversarial attacks (TAAs) demonstrate strong transferability when deceiving classification models. However, existing TAAs often perform unsatisfactorily or even fail when applied to structured tasks such as semantic segmentation and object detection. Encouragingly, recent studies that categorize transformations into non-spatial and spatial transformations inspire us to address this challenge. We find that for non-structured tasks, labels are spatially non-structured, and thus TAAs are not required to adjust labels when applying spatial transformations. In contrast, for structured tasks, labels are spatially structured, and failing to transform labels synchronously with inputs can cause spatial misalignment and yield erroneous gradients. To address these issues, we propose a novel unified Spatial Alignment Framework (SAF) for highly transferable TAAs on spatially structured tasks, where the TAAs spatially transform labels synchronously with the input using the proposed Spatial Alignment (SA) algorithm. Extensive experiments demonstrate the crucial role of our SAF for TAAs on structured tasks. Specifically, in non-targeted attacks, our SAF degrades the average mIoU on Cityscapes from 24.50 to 11.34, and on Kvasir-SEG from 49.91 to 31.80, while reducing the average mAP of COCO from 17.89 to 5.25.