Seg-XRes-CAM: Explaining Spatially Local Regions in Image Segmentation

Syed Nouman Hasany, Caroline Petitjean, Fabrice Mériaudeau

CVPRW 2023 pp. 3733-3738

doi:10.1109/CVPRW59228.2023.00384 /cvprw/2023/hasany2023cvprw-segxrescam/

Abstract

While many post-hoc model interpretability techniques exist for image classification, image segmentation has not received the same attention. An extension of Grad-CAM, Seg-Grad-CAM was proposed as a local interpretability technique for image segmentation. In this paper, we highlight that by virtue of its design, Seg-Grad-CAM does not utilize spatial information when it comes to generating explanations for regions within a segmentation map. Taking inspiration from HiResCAM, we propose Seg-XRes-CAM in order to solve this problem. We verify the utility of our proposed method by visually comparing explanations generated from Seg-Grad-CAM and Seg-XRes-CAM against a model-agnostic, perturbation-based method, RISE. The code is available at https://github.com/Nouman97/Seg_XRes_CAM.

CVPRW Semantic Scholar

Cite

Text

Hasany et al. "Seg-XRes-CAM: Explaining Spatially Local Regions in Image Segmentation." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2023. doi:10.1109/CVPRW59228.2023.00384

Markdown

[Hasany et al. "Seg-XRes-CAM: Explaining Spatially Local Regions in Image Segmentation." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2023.](https://mlanthology.org/cvprw/2023/hasany2023cvprw-segxrescam/) doi:10.1109/CVPRW59228.2023.00384

BibTeX

@inproceedings{hasany2023cvprw-segxrescam,
  title     = {{Seg-XRes-CAM: Explaining Spatially Local Regions in Image Segmentation}},
  author    = {Hasany, Syed Nouman and Petitjean, Caroline and Mériaudeau, Fabrice},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
  year      = {2023},
  pages     = {3733-3738},
  doi       = {10.1109/CVPRW59228.2023.00384},
  url       = {https://mlanthology.org/cvprw/2023/hasany2023cvprw-segxrescam/}
}