EUVIMEDEUVIMED entdecken
Crossref

OLMD: Orientation-aware Long-term Motion Decoupling for Continuous Sign Language Recognition

Yiheng Yu, Sheng Liu, Yuan Feng, Min Xu, Zhelun Jin, Xuhua Yang

Proceedings of the AAAI Conference on Artificial Intelligence · 11.04.2025 · 39 (9) · 9707-9715

DOI: 10.1609/aaai.v39i9.33052 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 11.04.2025. Verfasst von Yiheng Yu und 5 weiteren Autor:innen. Die Arbeit wurde bisher 8-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The primary challenge in continuous sign language recognition (CSLR) mainly stems from the presence of multi-orientational and long-term motions. However, current research overlooks these crucial aspects, significantly impacting accuracy. To tackle these issues, we propose a novel CSLR framework: Orientation-aware Long-term Motion Decoupling (OLMD), which efficiently aggregates long-term motions and decouples multi-orientational signals into easily interpretable components. Specifically, our innovative Long-term Motion Aggregation (LMA) module filters out static redundancy while adaptively capturing abundant features of long-term motions. We further enhance orientation awareness by decoupling complex movements into horizontal and vertical components, allowing for motion purification in both orientations. Additionally, two coupling mechanisms are proposed: stage and cross-stage coupling, which together enrich multi-scale features and improve the generalization capabilities of the model. Experimentally, OLMD shows SOTA performance on three large-scale datasets: PHOENIX14, PHOENIX14-T, and CSL-Daily. Notably, we improve the word error rate (WER) on PHOENIX14 by an absolute 1.6% compared to the previous SOTA.

Zitieren

APA

Yu, Y., Liu, S., Feng, Y., Xu, M., Jin, Z., & Yang, X. (2025). OLMD: Orientation-aware Long-term Motion Decoupling for Continuous Sign Language Recognition. Proceedings of the AAAI Conference on Artificial Intelligence, 39(9), 9707-9715. https://doi.org/10.1609/aaai.v39i9.33052

Vancouver

Yu Y, Liu S, Feng Y, Xu M, Jin Z, Yang X. OLMD: Orientation-aware Long-term Motion Decoupling for Continuous Sign Language Recognition. Proceedings of the AAAI Conference on Artificial Intelligence. 2025;39(9):9707-9715. doi:10.1609/aaai.v39i9.33052

BibTeX
@article{Yu2025,
  author = {Yu, Yiheng and Liu, Sheng and Feng, Yuan and Xu, Min and Jin, Zhelun and Yang, Xuhua},
  title = {OLMD: Orientation-aware Long-term Motion Decoupling for Continuous Sign Language Recognition},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2025},
  volume = {39},
  number = {9},
  pages = {9707--9715},
  doi = {10.1609/aaai.v39i9.33052}
}
Kostenlos starten →