arXiv cs.AI· Pushp Garg, Himangshu Sarma, Mrinmoy Ghorai·· 3 天前AI 评分22
混合运动学 GCN 与弹性 Transformer 实现高效印度手语识别
A Hybrid Kinematic GCN and Elastic Transformer for Efficient Indian Sign Language Recognition
AI 导读
研究提出一种结构感知混合框架用于印度手语（ISL）词汇级识别，结合熵驱动 SVD 采样器提取关键帧、连续双位置编码与距离感知 Elastic Transformer，以及运动学条件自适应 GCN 建模双手几何。团队构建了 120 类、9,453 段视频的 ISL 数据集，配合 Physical Node Dropout 与 ArcFace 优化，取得 89.8% 准确率，为复杂手语建模建立强基线。
正文 · 原文
Abstract:Sign Language Recognition (SLR) presents complex challenges due to variable signing speeds and spatial intricacies. Specifically, Indian Sign Language (ISL) requires precise modeling of its inherently bimanual nature and is bottlenecked by the scarcity of large-vocabulary benchmarks and the inability of standard models to handle irregular signing velocities. To address this, a structurally-aware hybrid framework for efficient, gloss ISL recognition is proposed. To minimize computational redundancy, an entropy-driven SVD sampler that dynamically extracts informative keyframes is used. Because this subsampling creates irregular physical time gaps, the proposed architecture employs a Continuous Dual Positional Encoding paired with a distance-aware Elastic Transformer to preserve natural signing rhythm. Spatially, a Kinematically-Conditioned Adaptive GCN explicitly models bimanual geometry using unified bone-joint features. To validate our framework, a 120-class, 9,453-video native ISL dataset is created. Augmented by Physical Node Dropout and ArcFace optimization to mitigate subject-specific memorization, our framework achieves a competitive 89.8% accuracy, establishing a strong baseline for complex sign language modeling.
| Subjects: | Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2610.14680 [cs.AI] |
| (or arXiv:2610.14680v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.14680 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Pushp Garg [view email]
[v1]
Sat, 10 Oct 2026 07:50:57 UTC (2,938 KB)
来源：arXiv cs.AI · arxiv.org