# A Hybrid Kinematic GCN and Elastic Transformer for Efficient Indian Sign Language Recognition

arXiv: https://arxiv.org/abs/2610.14680 · HTML: https://mlsanity.com/paper/2610.14680/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 20 of its 65 references that are in the map.
- First posted on arXiv 10 Oct 2026.

## Builds on (most cited first, 20 of 20)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training](https://mlsanity.com/paper/2203.12602/) (2022)
- [Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition](https://mlsanity.com/paper/1801.07455/) (2018)
- [V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning](https://mlsanity.com/paper/2506.09985/) (2025)
- [MViTv2: Improved Multiscale Vision Transformers for Classification and Detection](https://mlsanity.com/paper/2112.01526/) (2021)
- [Channel-wise Topology Refinement Graph Convolution for Skeleton-Based Action Recognition](https://mlsanity.com/paper/2107.12213/) (2021)
- [Word-level Deep Sign Language Recognition from Video: A New Large-scale Dataset and Methods Comparison](https://mlsanity.com/paper/1910.11006/) (2019)
- [RTMPose: Real-Time Multi-Person Pose Estimation based on MMPose](https://mlsanity.com/paper/2303.07399/) (2023)
- [Progressive Transformers for End-to-End Sign Language Production](https://mlsanity.com/paper/2004.14874/) (2020)
- [Including Signed Languages in Natural Language Processing](https://mlsanity.com/paper/2105.05222/) (2021)
- [ContiFormer: Continuous-Time Transformer for Irregular Time Series Modeling](https://mlsanity.com/paper/2402.10635/) (2024)
- [TGL: A General Framework for Temporal GNN Training on Billion-Scale Graphs](https://mlsanity.com/paper/2203.14883/) (2022)
- [OpenHands: Making Sign Language Recognition Accessible with Pose-based Pretrained Models across Languages](https://mlsanity.com/paper/2110.05877/) (2021)
- [Real-Time Sign Language Detection using Human Pose Estimation](https://mlsanity.com/paper/2008.04637/) (2020)
- [Considerations for meaningful sign language machine translation based on glosses](https://mlsanity.com/paper/2211.15464/) (2022)
- [Machine Translation between Spoken Languages and Signed Languages Represented in SignWriting](https://mlsanity.com/paper/2210.05404/) (2022)
- [WLASL-LEX: a Dataset for Recognising Phonological Properties in American Sign Language](https://mlsanity.com/paper/2203.06096/) (2022)
- [iSign: A Benchmark for Indian Sign Language Processing](https://mlsanity.com/paper/2407.05404/) (2024)
- [ISLTranslate: Dataset for Translating Indian Sign Language](https://mlsanity.com/paper/2307.05440/) (2023)
- [Sign Languague Recognition without frame-sequencing constraints: A proof of concept on the Argentinian Sign Language](https://mlsanity.com/paper/2310.17437/) (2023)

## Abstract (arXiv)

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.