Cross-lingual few-shot sign language recognition

Yunus Can Bilge, Nazli Ikizler-Cinbis, Ramazan Gokberk Cinbis

Abstract: Highlights•The motivation of the problem is to recognize a novel sign based on a small set of examples.•A novel framework leverages signer body and hand features for embedding is proposed.•Three novel meta-learning benchmarks that span multiple languages are introduced.•Our embedding framework achieves the best performance in three proposed benchmarks.