A Depth-Based Lightweight 3-D Dynamic Hand Gesture Recognition Framework

Qing Gao

doi:10.1109/tcds.2025.3617793

A Depth-Based Lightweight 3-D Dynamic Hand Gesture Recognition Framework

2025-10-08

0

PRE

AI

DOI：10.1109/tcds.2025.3617793

Original

Original request for help

Save

Abstract

En 中文

Hand gesture recognition is a key technology in the field of human–robot interaction (HRI). This article proposes a depth-based lightweight dynamic hand gesture recognition framework for HRI, which includes a 3-D hand pose estimation network based on biogeometry constraints (BCPoseNet) and a dynamic hand gesture recognition network based on multistream residual attention (MRAGesNet). BCPoseNet leverages its lightweight multiflow hierarchy structure to extract and refine the skeletons of palm and five fingers, respectively. Meanwhile, a set of biogeometry loss functions are designed to further constrain the relative position of skeletons. MRAGesNet constructs a lightweight 1-D ConvNet baseline, on which an adaptive cross-channel residual (ACR) block and multichannel attention mechanism (MAM) are introduced to explore the linkage mechanisms of gesture semantics and further enhance recognition performance. A feature calculation component is designed which enables a seamless integration of the framework through the transformation of four functions. The proposed methods achieve the state-of-the-art performance in accuracy and speed that are evaluated extensively on four public datasets and one custom dataset. In addition, a hand gesture imitation experiment on a hand–arm robot platform proves the application performance of the proposed framework for HRI.

Keywords:

Framework

hand gesture recognition

human–robot interaction (HRI)

lightweight structure

A Depth-Based Lightweight 3-D Dynamic Hand Gesture Recognition Framework

Abstract

Journal

Organization