Segat is a Pytorch project that provides a variety of FGSM attacks for semantic segmentation networks. The module contains an example on the popular architecture Deeplab V3.
- segat.py - Pytorch module consisting of various FGSM based attacks for semantic segmentation architectures
- cocoproc - Some preprocessing tools for the COCO dataset used in the example attack.
The code is distributed under a Creative Commons Attribution 4.0 International Public License. If you use this code please attribute to L. Stewart and P. Perez