Regional Multi-Scale Approach for Visually Pleasing Explanations of Deep Neural Networks
By Dasom Seo, Kanghan Oh, Il-Seok Oh. 2018
29 citations recorded by the source corpus.
Recently, many methods to interpret and visualize deep neural network predictions have been proposed, and significant progress has been made. However, a more class-discriminative and visually pleasing explanation is required. Thus, this paper proposes a region-based approach that estimates feature importance in terms of appropriately segmented regions. By fusing the saliency maps generated from multi-scale segmentations, a more class-discriminative and visually pleasing map is obtained. This paper incorporates this regional multi-scale concept into a prediction difference method that is model-agnostic. An input image is segmented in several scales using the superpixel method, and exclusion of a region is simulated by sampling a normal distribution constructed via the boundary prior. The experimental results demonstrate that the regional multi-scale method produces much more class-discriminative and visually pleasing saliency maps.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Lawrence Carin
- Trevor Darrell
- Marcus Rohrbach
- Chunyuan Li
- Sayna Ebrahimi
- Ricardo Henao
- Joseph E. Gonzalez
- Zhe Gan
In Computer Vision and Pattern Recognition
- Lawrence Carin
- Trevor Darrell
- Chunyuan Li
- Luc Van Gool
- Marcus Rohrbach
- Sayna Ebrahimi
- Joseph E. Gonzalez
- Dacheng Tao
Citations in the corpus
- Remembering for the Right Reasons: Explanations Reduce Catastrophic Forgetting (2020, methodology): “In contrast, perturbation-based methods are model-agnostic and produce saliency maps by observing the change in prediction when the input is perturbed (Petsiuk et al., 2018; Ribeiro et al., 2016; Ross et al., 2017; Zhou et al., 2014; Seo et al., 2018).”
- Proactive Pseudo-Intervention: Causally Informed Contrastive Learning For Interpretable Vision Models (2020, methodology): “Several methods follow a similar idea, but use other importance measures or occlusion strategies [43, 45, 50].”