How do Foundation Models Compare to Skeleton-Based Approaches for Gesture Recognition in Human-Robot Interaction?

TL;DR

This paper compares traditional skeleton-based gesture recognition methods with modern vision foundation and multimodal models, introducing a new dataset and analyzing their effectiveness for human-robot interaction in complex environments.

Contribution

It evaluates the performance of VFM and VLM models for gesture recognition, introduces the NUGGET dataset, and explores the potential of VFMs to simplify gesture recognition systems.

Findings

HD-GCN achieves the best performance among methods.

V-JEPA performs nearly as well with a simpler setup.

Gemini struggles with zero-shot gesture differentiation.

Abstract

Gestures enable non-verbal human-robot communication, especially in noisy environments like agile production. Traditional deep learning-based gesture recognition relies on task-specific architectures using images, videos, or skeletal pose estimates as input. Meanwhile, Vision Foundation Models (VFMs) and Vision Language Models (VLMs) with their strong generalization abilities offer potential to reduce system complexity by replacing dedicated task-specific modules. This study investigates adapting such models for dynamic, full-body gesture recognition, comparing V-JEPA (a state-of-the-art VFM), Gemini Flash 2.0 (a multimodal VLM), and HD-GCN (a top-performing skeleton-based approach). We introduce NUGGET, a dataset tailored for human-robot communication in intralogistics environments, to evaluate the different gesture recognition approaches. In our experiments, HD-GCN achieves best performance, but V-JEPA comes close with a simple, task-specific classification head - thus paving a possible way towards reducing system complexity, by using it as a shared multi-task model. In contrast, Gemini struggles to differentiate gestures based solely on textual descriptions in the zero-shot setting, highlighting the need of further research on suitable input representations for gestures.