Deep CNNs have been pushing the frontier of visual recognition over past years. Besides recognition accuracy, strong demands in understanding deep CNNs in the research community motivate developments of tools to dissect pre-trained models to visualize how they make predictions. Recent works further push the interpretability in the network learning stage to learn more meaningful representations. In this work, focusing on a specific area of visual recognition, we report our efforts towards interpretable face recognition. We propose a spatial activation diversity loss to learn more structured face representations. By leveraging the structure, we further design a feature activation diversity loss to push the interpretable representations to be discriminative and robust to occlusions. We demonstrate on three face recognition benchmarks that our proposed method is able to achieve the state-of-art face recognition accuracy with easily interpretable face representations.
Towards Interpretable Face Recognition
Bangjie Yin,Luan Tran,Haoxiang Li,Xiaohui Shen,Xiaoming Liu
Published 2018 in IEEE International Conference on Computer Vision
ABSTRACT
PUBLICATION RECORD
- Publication year
2018
- Venue
IEEE International Conference on Computer Vision
- Publication date
2018-05-02
- Fields of study
Computer Science
- Identifiers
- External record
- Source metadata
Semantic Scholar
CITATION MAP
EXTRACTION MAP
CLAIMS
CONCEPTS
- face recognition benchmarks
The evaluation datasets used to measure face recognition performance in the reported experiments.
- feature activation diversity loss
A loss term that encourages diversity at the feature-activation level to shape the learned face representation.
- interpretable face recognition
A face recognition setting focused on models whose internal representations are easy to visualize and understand.
- occlusions
Partial masking or blocking of a face by objects or other regions in the input image.
- spatial activation diversity loss
A loss term that encourages diverse spatial activation patterns in face representations.
- structured face representations
Face feature representations organized into more interpretable structure for analysis and recognition.
REFERENCES
Showing 1-75 of 75 references · Page 1 of 1
CITED BY
Showing 1-91 of 91 citing papers · Page 1 of 1