RoCoISLR: A Romanian Corpus for Isolated Sign Language Recognition

TL;DR

This paper introduces RoCoISLR, a new Romanian sign language dataset with over 9,000 videos, and benchmarks seven models, revealing the potential of transformer architectures for sign language recognition.

Contribution

The paper presents the first large-scale, standardized Romanian sign language dataset and provides benchmark evaluations of state-of-the-art models for isolated sign language recognition.

Findings

Transformer models outperform convolutional models

Swin Transformer achieved 34.1% Top-1 accuracy

Challenges due to long-tail class distributions in low-resource sign languages

Abstract

Automatic sign language recognition plays a crucial role in bridging the communication gap between deaf communities and hearing individuals; however, most available datasets focus on American Sign Language. For Romanian Isolated Sign Language Recognition (RoISLR), no large-scale, standardized dataset exists, which limits research progress. In this work, we introduce a new corpus for RoISLR, named RoCoISLR, comprising over 9,000 video samples that span nearly 6,000 standardized glosses from multiple sources. We establish benchmark results by evaluating seven state-of-the-art video recognition models-I3D, SlowFast, Swin Transformer, TimeSformer, Uniformer, VideoMAE, and PoseConv3D-under consistent experimental setups, and compare their performance with that of the widely used WLASL2000 corpus. According to the results, transformer-based architectures outperform convolutional baselines; Swin Transformer achieved a Top-1 accuracy of 34.1%. Our benchmarks highlight the challenges associated with long-tail class distributions in low-resource sign languages, and RoCoISLR provides the initial foundation for systematic RoISLR research.