# DYAN: A Dynamical Atoms-Based Network for Video Prediction

arXiv: https://arxiv.org/abs/1803.07201 · HTML: https://mlsanity.com/paper/1803.07201/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Mar 2018.

## Built on by (most cited first, 11 of 11)

- [Iso-Dream: Isolating and Leveraging Noncontrollable Visual Dynamics in World Models](https://mlsanity.com/paper/2205.13817/) (2022)
- [A Dynamic Multi-Scale Voxel Flow Network for Video Prediction](https://mlsanity.com/paper/2303.09875/) (2023)
- [Optimizing Video Prediction via Video Frame Interpolation](https://mlsanity.com/paper/2206.13454/) (2022)
- [Comparing Correspondences: Video Prediction with Correspondence-wise Losses](https://mlsanity.com/paper/2104.09498/) (2021)
- [Model-Based Reinforcement Learning with Isolated Imaginations](https://mlsanity.com/paper/2303.14889/) (2023)
- [Predicting Physical World Destinations for Commands Given to Self-Driving Cars](https://mlsanity.com/paper/2112.05419/) (2021)
- [Space-Time Forecasting of Dynamic Scenes with Motion-aware Gaussian Grouping](https://mlsanity.com/paper/2602.21668/) (2026)
- [FoV-Net: Field-of-View Extrapolation Using Self-Attention and Uncertainty](https://mlsanity.com/paper/2204.01267/) (2022)
- [Cross-view Action Recognition via Contrastive View-invariant Representation](https://mlsanity.com/paper/2305.01733/) (2023)
- [Future Object Detection with Spatiotemporal Transformers](https://mlsanity.com/paper/2204.10321/) (2022)
- [Lightweight Improved Residual Network for Efficient Inverse Tone Mapping](https://mlsanity.com/paper/2307.03998/) (2023)

## Abstract (arXiv)

The ability to anticipate the future is essential when making real time critical decisions, provides valuable information to understand dynamic natural scenes, and can help unsupervised video representation learning. State-of-art video prediction is based on LSTM recursive networks and/or generative adversarial network learning. These are complex architectures that need to learn large numbers of parameters, are potentially hard to train, slow to run, and may produce blurry predictions. In this paper, we introduce DYAN, a novel network with very few parameters and easy to train, which produces accurate, high quality frame predictions, significantly faster than previous approaches. DYAN owes its good qualities to its encoder and decoder, which are designed following concepts from systems identification theory and exploit the dynamics-based invariants of the data. Extensive experiments using several standard video datasets show that DYAN is superior generating frames and that it generalizes well across domains.