LNeRV: Disentangled Latent Representation for Semantic-Aware Video Dynamics Modeling
IEEE Transactions on Circuits and Systems for Video Technology · 2026-09 · 36 (9) · 13063-13073
DOI: 10.1109/tcsvt.2026.3698339 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Circuits and Systems for Video Technology, 2026-09. Verfasst von Yan Shen und 5 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 65 Literaturangaben. Gefördert von: National Key Research and Development Program of China, National Natural Science Foundation of China, Fundamental and Interdisciplinary Disciplines Breakthrough Plan of the Ministry of Education of China.
Abstract (Originalsprache)
Implicit Neural Representations (INRs) have emerged as a promising paradigm for video coding by enabling continuous signal modeling through coordinate-based neural networks. However, existing video INR methods often model temporal variation either in the pixel domain or with a single dynamic code, which limits their ability to jointly preserve rapidly varying high-frequency details and long-range semantic motion patterns. To address this issue, we propose the Latent-disentangled Neural Representation for Videos (LNeRV), a framework that models video dynamics in latent space using dual dynamic feature grids. Specifically, a denser high-resolution grid is used to capture short-term detail dynamics, while a coarser low-rate grid is used to capture long-range semantic motion patterns. On top of these complementary grids, we introduce a Clustering-based Context Aggregation (CCA) module to combine short-term motion details with long-range temporal context, and a Decoupled Gated Attention (DGA) module as a lightweight gated fusion mechanism to enlarge the receptive field during feature interaction. Extensive experiments demonstrate that LNeRV achieves state-of-the-art video regression performance on the UVG dataset under both PSNR and MS-SSIM metrics. Compared with vanilla NeRV, LNeRV achieves an average PSNR gain of 2.52 dB on UVG, while also showing strong structural fidelity in the compressed rate-distortion setting.
Förderung
- National Key Research and Development Program of China
- National Natural Science Foundation of China
- Fundamental and Interdisciplinary Disciplines Breakthrough Plan of the Ministry of Education of China
- Suzhou Key Core Technology Project
Zitieren
APA
Shen, Y., Man, H., Cui, W., Zuo, W., Fan, X., & Zhao, D. (2026). LNeRV: Disentangled Latent Representation for Semantic-Aware Video Dynamics Modeling. IEEE Transactions on Circuits and Systems for Video Technology, 36(9), 13063-13073. https://doi.org/10.1109/tcsvt.2026.3698339
Vancouver
Shen Y, Man H, Cui W, Zuo W, Fan X, Zhao D. LNeRV: Disentangled Latent Representation for Semantic-Aware Video Dynamics Modeling. IEEE Transactions on Circuits and Systems for Video Technology. 2026;36(9):13063-13073. doi:10.1109/tcsvt.2026.3698339
BibTeX
@article{Shen2026,
author = {Shen, Yan and Man, Hengyu and Cui, Wenxue and Zuo, Wangmeng and Fan, Xiaopeng and Zhao, Debin},
title = {LNeRV: Disentangled Latent Representation for Semantic-Aware Video Dynamics Modeling},
journal = {IEEE Transactions on Circuits and Systems for Video Technology},
year = {2026},
volume = {36},
number = {9},
pages = {13063--13073},
doi = {10.1109/tcsvt.2026.3698339}
}
EUVIMED