# Score-level Multi Cue Fusion for Sign Language Recognition

arXiv: https://arxiv.org/abs/2009.14139 · HTML: https://mlsanity.com/paper/2009.14139/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Sep 2020.

## Built on by (most cited first, 7 of 7)

- [Using Motion History Images with 3D Convolutional Networks in Isolated Sign Language Recognition](https://mlsanity.com/paper/2110.12396/) (2021)
- [A Comprehensive Review of Sign Language Recognition: Different Types, Modalities, and Datasets](https://mlsanity.com/paper/2204.03328/) (2022)
- [Transfer Learning for Cross-dataset Isolated Sign Language Recognition in Under-Resourced Datasets](https://mlsanity.com/paper/2403.14534/) (2024)
- [Sign and Search: Sign Search Functionality for Sign Language Lexica](https://mlsanity.com/paper/2107.13637/) (2021)
- [A Transformer-Based Multi-Stream Approach for Isolated Iranian Sign Language Recognition](https://mlsanity.com/paper/2407.09544/) (2024)
- [Sign language recognition based on deep learning and low-cost handcrafted descriptors](https://mlsanity.com/paper/2408.07244/) (2024)
- [SegSLR: Promptable Video Segmentation for Isolated Sign Language Recognition](https://mlsanity.com/paper/2509.10710/) (2025)

## Abstract (arXiv)

Sign Languages are expressed through hand and upper body gestures as well as facial expressions. Therefore, Sign Language Recognition (SLR) needs to focus on all such cues. Previous work uses hand-crafted mechanisms or network aggregation to extract the different cue features, to increase SLR performance. This is slow and involves complicated architectures. We propose a more straightforward approach that focuses on training separate cue models specializing on the dominant hand, hands, face, and upper body regions. We compare the performance of 3D Convolutional Neural Network (CNN) models specializing in these regions, combine them through score-level fusion, and use the weighted alternative. Our experimental results have shown the effectiveness of mixed convolutional models. Their fusion yields up to 19% accuracy improvement over the baseline using the full upper body. Furthermore, we include a discussion for fusion settings, which can help future work on Sign Language Translation (SLT).