Research map / Video / Video segmentation and tracking
Self-Attention ConvLSTM for Spatiotemporal Prediction
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| SA-ConvLSTM vs MIM* | ConvLSTM-based predictor · MovingMNIST | MAE: 101.1 -> 94.7 | Helped |
| SA-ConvLSTM vs MIM | ConvLSTM-based predictor · TaxiBJ | MSE: around 9.3% | Helped |
| SA-ConvLSTM vs PredRNN++ | ConvLSTM-based predictor · KTH | PSNR: 28.47 -> 29.33 | Helped |
| add self-attention to ConvLSTM (ablation) | ConvLSTM · MovingMNIST | MSE: 63.98 -> 58.25 | Helped |
| add self-attention to ConvLSTM (ablation) | ConvLSTM · TaxiBJ | MSE: 0.527 -> 0.410 | Helped |
| add extra memory M (ablation) | ConvLSTM · MovingMNIST | MSE: 63.98 -> 56.17 | Helped |
| add memory aggregation Zm (ablation) | SA-ConvLSTM · MovingMNIST | MSE: 55.60 -> 43.92 | Helped |
| add memory aggregation Zm (ablation) | SA-ConvLSTM · TaxiBJ | MSE: 0.408 -> 0.390 | Helped |
Closest papers by meaning
- Convolutional State Space Models for Long-Range Spatiotemporal Modeling (NeurIPS 2023 · similarity 0.89)
- Wavelet-Driven Spatiotemporal Predictive Learning: Bridging Frequency and Time Variations (AAAI 2024 · similarity 0.87)
- Temporal Attention Unit: Towards Efficient Spatiotemporal Predictive Learning (CVPR 2023 · similarity 0.87)
- OpenSTL: A Comprehensive Benchmark of Spatio-Temporal Predictive Learning (NeurIPS 2023 · similarity 0.87)
- ST-ReP: Learning Predictive Representations Efficiently for Spatial-Temporal Forecasting (AAAI 2025 · similarity 0.85)
- Causal Spatio-Temporal Prediction: An Effective and Efficient Multi-Modal Approach (NeurIPS 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.