Can I read Bohacek Sign Pose-Based Transformer For Word-Level Sign Language Recognition WACVW 2022 Paper on EtoBox?

Bohacek Sign Pose-Based Transformer For Word-Level Sign Language Recognition WACVW 2022 Paper by 1d sc is a document available to read on EtoBox.

What is Bohacek Sign Pose-Based Transformer For Word-Level Sign Language Recognition WACVW 2022 Paper about?

This paper presents a low-cost word-level sign language recognition system based on the Transformer model, focusing on pose estimation from 2D landmark locations. The authors introduce a robust pose normalization scheme and augmentations that significantly improve recognition accuracy, achieving state-of-the-art results on WLASL and LSA64 datasets. The proposed method, SPOTER, enhances the potential for mobile device applications in sign language recognition.

Author

1d sc

Language

EN