Papers › SwinLSTM: Improving Spatiotemporal Prediction Accuracy using Swin Transformer and LSTM

SwinLSTM: Improving Spatiotemporal Prediction Accuracy using Swin Transformer and LSTM

1 Jan 2023ICCV 2023 1archive 2025-07-28

Song Tang, Chuang Li, Pu Zhang, RongNian Tang

Integrating CNNs and RNNs to capture spatiotemporal dependencies is a prevalent strategy for spatiotemporal prediction tasks. However, the property of CNNs to learn local spatial information decreases their efficiency in capturing spatiotemporal dependencies, thereby limiting their prediction accuracy. In this paper, we propose a new recurrent cell, SwinLSTM, which integrates Swin Transformer blocks and the simplified LSTM, an extension that replaces the convolutional structure in ConvLSTM with the self-attention mechanism. Furthermore, we construct a network with SwinLSTM cell as the core for spatiotemporal prediction. Without using unique tricks, SwinLSTM outperforms state-of-the-art methods on Moving MNIST, Human3.6m, TaxiBJ, and KTH datasets. In particular, it exhibits a significant improvement in prediction accuracy compared to ConvLSTM. Our competitive experimental results demonstrate that learning global spatial dependencies is more advantageous for models to capture spatiotemporal dependencies. We hope that SwinLSTM can serve as a solid baseline to promote the advancement of spatiotemporal prediction accuracy. The codes are publicly available at https://github.com/SongTang-x/SwinLSTM.

PaperPDFCode

Code

SongTang-x/SwinLSTM officialmentioned in paperpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

PredictionVideo Prediction

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Video Prediction Human3.6M SwinLSTM MAE 1190 #2 of 9 Archive leaderboard report
Video Prediction Human3.6M SwinLSTM MSE 332 #2 of 9 Archive leaderboard report
Video Prediction Human3.6M SwinLSTM SSIM 0.913 #2 of 9 Archive leaderboard report
Video Prediction Moving MNIST SwinLSTM MSE 17.7 #8 of 31 Archive leaderboard report
Video Prediction Moving MNIST SwinLSTM SSIM 0.962 #8 of 31 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Absolute Position EncodingsAdamAttentionBPEConvLSTMConvolutionDense ConnectionsDropoutLSTMLabel SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSigmoid ActivationSoftmaxStochastic DepthSwin TransformerTanh ActivationTransformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections