LA-Sign: Looped Transformers with Geometry-aware Alignment for Skeleton-based Sign Language Recognition
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866918495378210816 |
|---|---|
| author | Pu, Muxin Lim, Mei Kuan Chong, Chun Yong Loy, Chen Change |
| author_facet | Pu, Muxin Lim, Mei Kuan Chong, Chun Yong Loy, Chen Change |
| contents | Skeleton-based isolated sign language recognition (ISLR) demands fine-grained understanding of articulated motion across multiple spatial scales, from subtle finger movements to global body dynamics. Existing approaches typically rely on deep feed-forward architectures, which increase model capacity but lack mechanisms for recurrent refinement and structured representation. We propose LA-Sign, a looped transformer framework with geometry-aware alignment for ISLR. Instead of stacking deeper layers, LA-Sign derives its depth from recurrence, repeatedly revisiting latent representations to progressively refine motion understanding under shared parameters. To further regularise this refinement process, we present a geometry-aware contrastive objective that projects skeletal and textual features into an adaptive hyperbolic space, encouraging multi-scale semantic organisation. We study three looping designs and multiple geometric manifolds, demonstrating that encoder-decoder looping combined with adaptive Poincare alignment yields the strongest performance. Extensive experiments on WLASL and MSASL benchmarks show that LA-Sign achieves state-of-the-art results while using fewer unique layers, highlighting the effectiveness of recurrent latent refinement and geometry-aware representation learning for sign language recognition. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_29057 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | LA-Sign: Looped Transformers with Geometry-aware Alignment for Skeleton-based Sign Language Recognition Pu, Muxin Lim, Mei Kuan Chong, Chun Yong Loy, Chen Change Computer Vision and Pattern Recognition Skeleton-based isolated sign language recognition (ISLR) demands fine-grained understanding of articulated motion across multiple spatial scales, from subtle finger movements to global body dynamics. Existing approaches typically rely on deep feed-forward architectures, which increase model capacity but lack mechanisms for recurrent refinement and structured representation. We propose LA-Sign, a looped transformer framework with geometry-aware alignment for ISLR. Instead of stacking deeper layers, LA-Sign derives its depth from recurrence, repeatedly revisiting latent representations to progressively refine motion understanding under shared parameters. To further regularise this refinement process, we present a geometry-aware contrastive objective that projects skeletal and textual features into an adaptive hyperbolic space, encouraging multi-scale semantic organisation. We study three looping designs and multiple geometric manifolds, demonstrating that encoder-decoder looping combined with adaptive Poincare alignment yields the strongest performance. Extensive experiments on WLASL and MSASL benchmarks show that LA-Sign achieves state-of-the-art results while using fewer unique layers, highlighting the effectiveness of recurrent latent refinement and geometry-aware representation learning for sign language recognition. |
| title | LA-Sign: Looped Transformers with Geometry-aware Alignment for Skeleton-based Sign Language Recognition |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.29057 |