The author presents LAFS, a landmark-based self-supervised learning approach, to enhance face recognition performance by leveraging facial landmarks and self-supervised pretraining.
自己教育学習を活用した顔認識のためのランドマークベースの手法による効果的な表現学習