Temporal Attention Unit: Towards Efficient Spatiotemporal Predictive Learning
Spatiotemporal predictive learning aims to generate future frames by learning from historical frames. In this paper, we investigate existing methods and present a general framework of spatiotemporal predictive learning, in which the spatial encoder and decoder capture intra-frame features and the middle temporal module catches inter-frame correlations. While the mainstream methods employ recurrent units to capture long-term temporal dependencies, they suffer from low computational efficiency due to their unparallelizable architectures. To parallelize the temporal module, we propose the Temporal Attention Unit (TAU), which decomposes the temporal attention into intra-frame statical attention and inter-frame dynamical attention. Moreover, while the mean squared error loss focuses on intra-frame errors, we introduce a novel differential divergence regularization to take inter-frame variations into account. Extensive experiments demonstrate that the proposed method enables the derived model to achieve competitive performance on various spatiotemporal prediction benchmarks.
Code (2)
Tasks
Computational EfficiencyDecoderVideo PredictionSimilar Papers 제목 키워드 기반
Triplet Attention Transformer for Spatiotemporal Predictive Learning
Spatiotemporal predictive learning offers a self-supervised learning paradigm that enables models to learn both spatial and temporal patterns by predicting future sequences based on historical sequences. Mainstream metho…
Computational EfficiencyPredictionSelf-Supervised LearningTrajectory Prediction+1PredRNN: Recurrent Neural Networks for Predictive Learning using Spatiotemporal LSTMs
The predictive learning of spatiotemporal sequences aims to generate future images by learning from the historical frames, where spatial appearances and temporal variations are two crucial structures. This paper models t…
Video PredictionPredRNN: Recurrent Neural Networks for Predictive Learning using Spatiotemporal LSTMs
The predictive learning of spatiotemporal sequences aims to generate future images by learning from the historical frames, where spatial appearances and temporal variations are two crucial structures. This paper models t…
Video PredictionSimVPv2: Towards Simple yet Powerful Spatiotemporal Predictive Learning
Recent years have witnessed remarkable advances in spatiotemporal predictive learning, with methods incorporating auxiliary inputs, complex neural architectures, and sophisticated training strategies. While SimVP has int…
Computational EfficiencyVideo PredictionA Gravity-informed Spatiotemporal Transformer for Human Activity Intensity Prediction
Human activity intensity prediction is a crucial to many location-based services. Although tremendous progress has been made to model dynamic spatiotemporal patterns of human activity, most existing methods, including sp…