# Video Frame Interpolation via Structure-Motion based Iterative Fusion

arXiv: https://arxiv.org/abs/2105.05353 · HTML: https://mlsanity.com/paper/2105.05353/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 16 references that are in the map.
- First posted on arXiv 11 May 2021.

## Builds on (most cited first, 9 of 9)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Multi-Scale Context Aggregation by Dilated Convolutions](https://mlsanity.com/paper/1511.07122/) (2015)
- [PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volume](https://mlsanity.com/paper/1709.02371/) (2017)
- [EDVR: Video Restoration with Enhanced Deformable Convolutional Networks](https://mlsanity.com/paper/1905.02716/) (2019)
- [U2-Net: Going Deeper with Nested U-Structure for Salient Object Detection](https://mlsanity.com/paper/2005.09007/) (2020)
- [Video Frame Interpolation via Adaptive Separable Convolution](https://mlsanity.com/paper/1708.01692/) (2017)
- [AdaCoF: Adaptive Collaboration of Flows for Video Frame Interpolation](https://mlsanity.com/paper/1907.10244/) (2019)
- [MEMC-Net: Motion Estimation and Motion Compensation Driven Neural Network for Video Interpolation and Enhancement](https://mlsanity.com/paper/1810.08768/) (2018)

## Built on by (most cited first, 1 of 1)

- [SketchBetween: Video-to-Video Synthesis for Sprite Animation via Sketches](https://mlsanity.com/paper/2209.00185/) (2022)

## Abstract (arXiv)

Video Frame Interpolation synthesizes non-existent images between adjacent frames, with the aim of providing a smooth and consistent visual experience. Two approaches for solving this challenging task are optical flow based and kernel-based methods. In existing works, optical flow based methods can provide accurate point-to-point motion description, however, they lack constraints on object structure. On the contrary, kernel-based methods focus on structural alignment, which relies on semantic and apparent features, but tends to blur results. Based on these observations, we propose a structure-motion based iterative fusion method. The framework is an end-to-end learnable structure with two stages. First, interpolated frames are synthesized by structure-based and motion-based learning branches respectively, then, an iterative refinement module is established via spatial and temporal feature integration. Inspired by the observation that audiences have different visual preferences on foreground and background objects, we for the first time propose to use saliency masks in the evaluation processes of the task of video frame interpolation. Experimental results on three typical benchmarks show that the proposed method achieves superior performance on all evaluation metrics over the state-of-the-art methods, even when our models are trained with only one-tenth of the data other methods use.