SRENet: Spectral re-entry network for point cloud action recognition
Author Identifier (ORCID)
Abstract
Recognizing human actions from point cloud sequences is critical for 3D perception driven applications such as autonomous driving and human-computer interaction. However, the irregular structure and temporal inconsistency of point clouds pose unique challenges for spatio-temporal representation learning, especially in capturing both global motion context and fine-grained temporal dynamics. We propose SRENet, a spectral-aware framework designed to explicitly learn both global context and fine-grained temporal dynamics of motion from a frequency perspective for action recognition. SRENet introduces a Spectral Decomposition Block (SDeBlock) that performs wavelet-based analysis along temporal and spatial axes, disentangling features into low- and high-frequency components with frequency-specific attention. To recover residual dynamics and re-align temporal frequency structures distorted during semantic fusion, a Spectral Re-entry Block (SReBlock) performs secondary temporal decomposition. Furthermore, a spectral-aware learning strategy is devised to enhance discriminability in both frequency subspaces via contrastive loss and a curriculum schedule that gradually shifts focus from low- to high-frequency spaces in line with coarse to detailed motion patterns. Extensive experiments on MSR-Action3D, NTU-RGBD and NTU-RGBD120 demonstrate that SRENet achieves state-of-the-art performance, validating the effectiveness of frequency modeling in point cloud-based action understanding 1.
Keywords
action recognition, frequency modeling, point cloud video
Document Type
Journal Article
Date of Publication
1-1-2026
Publication Title
IEEE Transactions on Circuits and Systems for Video Technology
Publisher
IEEE
School
School of Science
Copyright
subscription content
Comments
Wu, Q., Lan, J., Kang, W., Wang, Z., & Hu, K. (2026). SRENet: Spectral re-entry network for point cloud action recognition. IEEE Transactions on Circuits and Systems for Video Technology. Advance online publication. https://doi.org/10.1109/TCSVT.2026.3695515