New lightweight Transformer model boosts sign language recognition accuracy

Researchers have developed TransSLR, a lightweight Transformer model designed for sign language recognition, particularly for under-represented languages like Central African Sign Language (CASL). This model operates on geometric keypoint representations rather than raw RGB images, enabling signer-independent generalization and reducing computational requirements. TransSLR achieves a new state-of-the-art accuracy of 80.39% on the CASL-W60 benchmark, surpassing previous methods by over 10%. AI

IMPACT This model could enable more accessible sign language recognition tools for under-represented languages, improving communication and inclusivity.

RANK_REASON The cluster contains an academic paper detailing a new model and benchmark results. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• CASL-W60
• Central African Sign Language
• Lucia Yen Wanchi
• TransSLR
• WLASL

• paper
• model release
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →