tsingqguo/anglerocl
We introduce Angle-Robust Concept Learning (AngleRoCL), a simple and flexible approach that learns a generalizable concept representing the capability of generating angle-robust patches. The learned concept can be incorporated into textual prompts and guides T2I to generate patches with their attack inherently resistant to viewpoint variations.
GitHub repository with 5 stars and 1 forks.
Language: Jupyter Notebook