DFDNet: Disentangling and Filtering Dynamics for Enhanced Video Prediction

Videos inherently contain complex temporal dynamics across various spatial directions, often entangled in ways that obscure effective dynamic extraction. Previous studies typically process video spatiotemporal features without disentangling, which hampers their ability to extract dynamic information. Additionally, the extraction of dynamics is disrupted by transient high-dynamic information in video sequences, e.g., noise or flicker, which has received limited attention in the literature. To tackle those problems, this paper proposes the Disentangling and Filtering Dynamics Network (DFD-Net). Firstly, to disentangle the interwoven dynamics, DFD-Net decomposes the spatially encoded video sequences into lower dimensional sequences. Secondly, a learnable threshold filter is proposed to eliminate the transient high-dynamic information. Thirdly, the model incorporates an MLP to extract the temporal dependencies from the disentangled and filtered sequences. DFDNet demonstrates competitive performance across four chosen datasets, including both low and highresolution videos. Specifically, on the low-resolution Moving MNIST dataset, DFDNet achieves a 19% improvement on MSE over the previous state-of-the-art model. On the highresolution SJTU4K dataset, it outperforms the previous stateof-the-art model by 10% on the LPIPS metric under similar inference time.