MHRGait: Gait Recognition from Momentum Human Rig Pose

2026-07-31

AI Summary

This work addresses the limitations of existing gait recognition methods that rely on geometric representations such as silhouettes or skeletons, which are highly susceptible to variations in clothing and body shape and often fail to capture identity-relevant joint motion characteristics. To overcome this, we propose MHRGait++, which, for the first time, leverages the 184-dimensional semantic joint control parameters from the Momentum Human Rig (MHR) pose as a standalone gait representation. Our approach models intra-frame coordination and temporal dynamics by grouping these parameters according to anatomical structure and introduces a modality-balanced distance fusion mechanism to effectively integrate silhouette information while preventing descriptor count from dominating modality weighting. Evaluated on CCPG and SUSTech1K, MHRGait++ surpasses current model-based methods and demonstrates significantly improved cross-dataset generalization. Notably, MHRGait achieves high accuracy with only 2.76M parameters and 0.69 GFLOPs for a 30-frame input, striking an optimal balance between performance and efficiency.

Abstract

Gait recognition is shaped by its input representation. Silhouettes encode projected body shape, skeletons encode sparse joint coordinates, and 3D meshes encode dense surface geometry. In each case, identity-bearing articulation is observed through geometric carriers that also vary with clothing, skeletal scale, or body shape. We investigate whether gait can instead be recognized from compact articulated controls. We introduce Momentum Human Rig (MHR) pose as a gait representation, describing each frame using 184 semantically organized body and hand parameters estimated from monocular video. MHRGait groups these heterogeneous controls by anatomy, models their intra-frame coordination and temporal evolution, and produces compact body and hand descriptors. We further introduce MHRGait++, which combines MHR pose with silhouettes through modality-balanced distance fusion, preventing descriptor count from determining modality importance. Experiments on four benchmarks show that MHRGait attains the best overall performance among compared model-based methods on CCPG and SUSTech1K and transfers effectively across datasets, while its recognition network requires only 2.76M parameters and 0.69 GFLOPs for a 30-frame input. MHRGait++ consistently improves silhouette recognizers with a favorable accuracy-efficiency trade-off. These results establish rig-space articulation as an effective standalone gait representation and a complementary cue to projected body shape. Our code is available at https://github.com/duanhuiran/MHRGait.

Problem

• gait recognition
• articulated controls
• input representation
• rig pose
• modality fusion

Innovation

• Momentum Human Rig
• gait representation
• articulated controls
• modality-balanced fusion
• efficient gait recognition