用频率引导的门控机制，让卷积网络自适应捕捉运动模式。
PFGNet: A Fully Convolutional Frequency-Guided Peripheral Gating Network for Efficient Spatiotemporal Predictive Learning
- 通过像素级频率引导门控，动态调节感受野大小。
- 在多个数据集上达到顶尖或接近顶尖性能，参数与计算量大幅减少。
- 适合追求高效时序预测的开发者，尤其关注轻量化模型设计。
时空预测学习（STPL）旨在从历史观测中预测未来帧，广泛应用于各类场景。相比循环或混合架构，纯卷积模型具有更高的效率和完全并行性，但固定感受野限制了其对空间变化运动模式的适应能力。受生物中心-周围结构和频率选择性信号处理启发，本文提出PFGNet，一种全卷积框架，通过像素级频率引导门控动态调制感受野。核心外围频率门控（PFG）模块提取局部频谱线索，自适应融合多尺度大核外围响应与可学习中心抑制，有效形成空间自适应带通滤波器。为保持高效性，所有大核均分解为可分离的一维卷积（1×k后接k×1），将每通道计算成本从O(k²)降至O(2k)。PFGNet实现无循环、无注意力的结构感知时空建模。在Moving MNIST、TaxiBJ、Human3.6M和KTH数据集上的实验表明，PFGNet在显著减少参数量和浮点运算次数的前提下，取得当前最优或接近最优的预测性能。代码已开源：https://github.com/fhjdqaq/PFGNet。
原文摘要 · Abstract (English)
Spatiotemporal predictive learning (STPL) aims to forecast future frames from past observations and is essential across a wide range of applications. Compared with recurrent or hybrid architectures, pure convolutional models offer superior efficiency and full parallelism, yet their fixed receptive fields limit their ability to adaptively capture spatially varying motion patterns. Inspired by biological center-surround organization and frequency-selective signal processing, we propose PFGNet, a fully convolutional framework that dynamically modulates receptive fields through pixel-wise frequency-guided gating. The core Peripheral Frequency Gating (PFG) block extracts localized spectral cues and adaptively fuses multi-scale large-kernel peripheral responses with learnable center suppression, effectively forming spatially adaptive band-pass filters. To maintain efficiency, all large kernels are decomposed into separable 1D convolutions ($1 \times k$ followed by $k \times 1$), reducing per-channel computational cost from $O(k^2)$ to $O(2k)$. PFGNet enables structure-aware spatiotemporal modeling without recurrence or attention. Experiments on Moving MNIST, TaxiBJ, Human3.6M, and KTH show that PFGNet delivers SOTA or near-SOTA forecasting performance with substantially fewer parameters and FLOPs. Our code is available at https://github.com/fhjdqaq/PFGNet.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。