We lifted 2 functions out of this paper's own repositories and ran 1 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| ZhengyuZhao/PerC-Adversarial | canonical | 1 of 1 |
| jeromerony/adversarial-library | extension | 0 of 1 |
| Function | Status | Where it lives |
|---|---|---|
| quantization | Ran | ZhengyuZhao/PerC-Adversarial/perc_cw.py code served (permissive licence) · get_code("404ede55dcc21fe2") |
| perc_al | Not yet run | jeromerony/adversarial-library/adv_lib/attacks/perceptual_color_attacks/perceptual_color_distance_al.py code served (permissive licence) · get_code("c56c9f8b3a348e0a") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
The success of image perturbations that are designed to fool image classifier is assessed in terms of both adversarial effect and visual imperceptibility. The conventional assumption on imperceptibility is that perturbations should strive for tight $L_p$-norm bounds in RGB space. In this work, we drop this assumption by pursuing an approach that exploits human color perception, and more specifically, minimizing perturbation size with respect to perceptual color distance. Our first approach, Perceptual Color distance C&W (PerC-C&W), extends the widely-used C&W approach and produces larger RGB perturbations. PerC-C&W is able to maintain adversarial strength, while contributing to imperceptibility. Our second approach, Perceptual Color distance Alternating Loss (PerC-AL), achieves the same outcome, but does so more efficiently by alternating between the classification loss and perceptual color difference when updating perturbations. Experimental evaluation shows PerC approaches outperform conventional $L_p$ approaches in terms of robustness and transferability, and also demonstrates that the PerC distance can provide added value on top of existing structure-based methods to creating image perturbations.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("1911.02466")
get_code_for_paper("1911.02466")
have("1911.02466")
Connect an agent — have() is free.