Score-level Multi Cue Fusion for Sign Language Recognition

TL;DR

This paper introduces a score-level fusion approach using specialized 3D CNN models for different sign language cues, significantly improving recognition accuracy over baseline methods.

Contribution

It presents a straightforward, multi-cue fusion method with separate models for hand, face, and upper body, enhancing sign language recognition performance.

Findings

Up to 19% accuracy improvement over baseline.

Effective fusion of specialized cue models.

Discussion on fusion settings for future SLT work.

Abstract

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).