# Effect of Architectures and Training Methods on the Performance of Learned Video Frame Prediction

arXiv: https://arxiv.org/abs/2008.06106 · HTML: https://mlsanity.com/paper/2008.06106/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Aug 2020.

## Built on by (most cited first, 4 of 4)

- [DFPN: Deformable Frame Prediction Network](https://mlsanity.com/paper/2105.12794/) (2021)
- [Mini-Batch Learning Strategies for modeling long term temporal dependencies: A study in environmental applications](https://mlsanity.com/paper/2210.08347/) (2022)
- [Message Propagation Through Time: An Algorithm for Sequence Dependency Retention in Time Series Modeling](https://mlsanity.com/paper/2309.16882/) (2023)
- [FG-DFPN: Flow Guided Deformable Frame Prediction Network](https://mlsanity.com/paper/2503.11343/) (2025)

## Abstract (arXiv)

We analyze the performance of feedforward vs. recurrent neural network (RNN) architectures and associated training methods for learned frame prediction. To this effect, we trained a residual fully convolutional neural network (FCNN), a convolutional RNN (CRNN), and a convolutional long short-term memory (CLSTM) network for next frame prediction using the mean square loss. We performed both stateless and stateful training for recurrent networks. Experimental results show that the residual FCNN architecture performs the best in terms of peak signal to noise ratio (PSNR) at the expense of higher training and test (inference) computational complexity. The CRNN can be trained stably and very efficiently using the stateful truncated backpropagation through time procedure, and it requires an order of magnitude less inference runtime to achieve near real-time frame prediction with an acceptable performance.