Improve face recognition by using image quality to mark hard examples with AdaFace
Improve face recognition by using image quality to mark hard examples with AdaFace
AdaFace: Quality Adaptive Margin for Face Recognition
arXiv paper abstract https://arxiv.org/abs/2204.00964v1
arXiv PDF paper https://arxiv.org/pdf/2204.00964v1.pdf
GitHub https://github.com/mk-minchul/adaface
Recognition in low quality face datasets is challenging because facial attributes are obscured and degraded.
… previous studies have studied … assign more importance to misclassified (hard) examples.
… argue … the relative importance of easy or hard samples should be based on the sample’s image quality.
… propose a new loss function that emphasizes samples of different difficulties based on their image quality.
… method achieves this in the form of an adaptive margin function by approximating the image quality with feature norms.
… show … AdaFace, improves the face recognition performance over the state-of-the-art (SoTA) on four datasets (IJB-B, IJB-C, IJB-S and TinyFace). …
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website
LinkedIn https://www.linkedin.com/in/morris-lee-47877b7b
