# Video Extrapolation in Space and Time

arXiv: https://arxiv.org/abs/2205.02084 · HTML: https://mlsanity.com/paper/2205.02084/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 27 of its 45 references that are in the map.
- First posted on arXiv 4 May 2022.

## Builds on (most cited first, 27 of 27)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [pixelNeRF: Neural Radiance Fields from One or Few Images](https://mlsanity.com/paper/2012.02190/) (2020)
- [D-NeRF: Neural Radiance Fields for Dynamic Scenes](https://mlsanity.com/paper/2011.13961/) (2020)
- [Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting](https://mlsanity.com/paper/1506.04214/) (2015)
- [Neural Scene Flow Fields for Space-Time View Synthesis of Dynamic Scenes](https://mlsanity.com/paper/2011.13084/) (2020)
- [Non-Rigid Neural Radiance Fields: Reconstruction and Novel View Synthesis of a Dynamic Scene From Monocular Video](https://mlsanity.com/paper/2012.12247/) (2020)
- [Space-time Neural Irradiance Fields for Free-Viewpoint Video](https://mlsanity.com/paper/2011.12950/) (2020)
- [SynSin: End-to-end View Synthesis from a Single Image](https://mlsanity.com/paper/1912.08804/) (2019)
- [Neural Radiance Flow for 4D View Synthesis and Video Processing](https://mlsanity.com/paper/2012.09790/) (2020)
- [Single-View View Synthesis with Multiplane Images](https://mlsanity.com/paper/2004.11364/) (2020)
- [DeepView: View Synthesis with Learned Gradient Descent](https://mlsanity.com/paper/1906.07316/) (2019)
- [Infinite Nature: Perpetual View Generation of Natural Scenes from a Single Image](https://mlsanity.com/paper/2012.09855/) (2020)
- [Video Frame Synthesis using Deep Voxel Flow](https://mlsanity.com/paper/1702.02463/) (2017)
- [3D Photography using Context-aware Layered Depth Inpainting](https://mlsanity.com/paper/2004.04727/) (2020)
- [PredRNN: A Recurrent Neural Network for Spatiotemporal Predictive Learning](https://mlsanity.com/paper/2103.09504/) (2021)
- [View Synthesis by Appearance Flow](https://mlsanity.com/paper/1605.03557/) (2016)
- [MINE: Towards Continuous Depth MPI with NeRF for Novel View Synthesis](https://mlsanity.com/paper/2103.14910/) (2021)
- [Video (language) modeling: a baseline for generative models of natural videos](https://mlsanity.com/paper/1412.6604/) (2014)
- [Layer-structured 3D Scene Inference via View Synthesis](https://mlsanity.com/paper/1807.10264/) (2018)
- [Worldsheet: Wrapping the World in a 3D Sheet for View Synthesis from a Single Image](https://mlsanity.com/paper/2012.09854/) (2020)
- [Future Video Synthesis with Object Motion Prediction](https://mlsanity.com/paper/2004.00542/) (2020)
- [Omnimatte: Associating Objects and Their Effects in Video](https://mlsanity.com/paper/2105.06993/) (2021)
- [Disentangling Propagation and Generation for Video Prediction](https://mlsanity.com/paper/1812.00452/) (2018)
- [Video Autoencoder: self-supervised disentanglement of static 3D structure and motion](https://mlsanity.com/paper/2110.02951/) (2021)
- [Learning Semantic-Aware Dynamics for Video Prediction](https://mlsanity.com/paper/2104.09762/) (2021)
- [NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis](https://mlsanity.com/paper/2003.08934/) (2020)
- [View Synthesis of Dynamic Scenes based on Deep 3D Mask Volume](https://mlsanity.com/paper/2108.13408/) (2021)

## Abstract (arXiv)

Novel view synthesis (NVS) and video prediction (VP) are typically considered disjoint tasks in computer vision. However, they can both be seen as ways to observe the spatial-temporal world: NVS aims to synthesize a scene from a new point of view, while VP aims to see a scene from a new point of time. These two tasks provide complementary signals to obtain a scene representation, as viewpoint changes from spatial observations inform depth, and temporal observations inform the motion of cameras and individual objects. Inspired by these observations, we propose to study the problem of Video Extrapolation in Space and Time (VEST). We propose a model that leverages the self-supervision and the complementary cues from both tasks, while existing methods can only solve one of them. Experiments show that our method achieves performance better than or comparable to several state-of-the-art NVS and VP methods on indoor and outdoor real-world datasets.