Color histogram equalization and fine-tuning to improve expression recognition of (partially occluded) faces on sign language datasets
Figure 1: Example pictures of our procedure to classify Ekman emotions.The original picture is first normalized in size, rotation, and color profile.Then, we perform classification on the full face, only top part, and only bottom part.On the left: examples from the AffectNet dataset.
Authors
- Fabrizio Nunnari (ORCID: https://orcid.org/0000-0002-1596-4043)
- Patrick Gebhard (ORCID: https://orcid.org/0000-0002-5566-4520)
Institutions
- German Research Centre for Artificial Intelligence (DE)
Publication Details
- Published
- 2025-09-16
- DOI
- https://doi.org/10.1145/3742886.3756731
- Primary Topic
- Hand Gesture Recognition Systems
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Bundesministerium für Bildung und Forschung