VideoMogaNet：面向视频识别的时空多阶交互方法
Spatio-Temporal Multi-Order Interaction for Video Recognition
AI 导读
研究者提出 VideoMogaNet，基于 MogaNet 的多阶空间表示改进视频识别。该方法用仅深度卷积与隐式门控聚合实现并行多阶时序处理，将复杂度从对通道数的二次方降为线性，并引入时空双交互以增强表示且不增加计算成本。在 UCF101、HMDB51、Kinetics-400、Diving48 等基准上，其性能达到或超过最先进动作识别方法，同时大幅降低时序分支开销。
正文 · 原文
Abstract:Many temporal approaches for video recognition rely on explicit cross-channel aggregation projections, which have quadratic complexity in the number of channels, introducing significant computational overhead. In addition, the spatio-temporal representation used for video recognition is biased either toward local order interactions (for CNNs) or toward global order interactions (for transformers). Starting from the multi-order spatial representation of MogaNet, our method introduces two complementary contributions to tackle these problems. First, for the temporal modelling, we introduce parallel multi-order processing with depthwise-only convolutions and implicit gating-based aggregation, which is linear in the number of channels, inducing a considerable computational reduction. Second, we propose a dual interaction between spatial and temporal dimensions, which allows a better spatio-temporal representation without increasing the computational cost. By combining these two contributions, our proposed VideoMogaNet achieves performance on par with or better than state-of-the-art action recognition approaches across multiple video benchmarks (UCF101, HMDB51, Kinetics-400, Diving48), while considerably reducing the cost of the temporal branch.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2611.05633 [cs.CV] |
| (or arXiv:2611.05633v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2611.05633 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Kevin Nguetche [view email]
[v1]
Tue, 3 Nov 2026 21:20:04 UTC (2,713 KB)
来源：arXiv cs.AI · arxiv.org