Support LSA64 Dataset

[balaboom123]

Add support for the LSA64 Dataset (Argentinian Sign Language) as detailed in Favaletta et al., 2023 in the modular preprocessing pipeline.

Background

LSA64 Dataset is a widely used Argentinian Sign Language resource that includes both video and pose data. Supporting it will expand the range and generalizability of the pipeline.

Requirements

• Integrate data loaders and converters for the LSA64 Dataset
• Enable LSA64 as a configurable dataset option in the pipeline configuration
• Ensure compatibility with the current preprocessing modules (MediaPipe, MMPose, YOLO)
• Document format, requirements, usage, and configuration differences in the documentation

Acceptance Criteria

• LSA64 data can be loaded and processed using the pipeline configuration
• Example scripts and config files for LSA64 are included and tested
• Documentation is updated to explain LSA64 dataset integration and use