Video Frame Interpolation via Adaptive Separable Convolution
Simon Niklaus, Long Mai, and Feng Liu
Computer Science Department
Portland State University
Abstract

Standard video frame interpolation methods first estimate optical flow between input frames and then synthesize an intermediate frame guided by motion. Recent approaches merge these two steps into a single convolution process by convolving input frames with spatially adaptive kernels that account for motion and re-sampling simultaneously. These methods require large kernels to handle large motion, which limits the number of pixels whose kernels can be estimated at once due to the large memory demand. To address this problem, this paper formulates frame interpolation as local separable convolution over input frames using pairs of 1D kernels. Compared to regular 2D kernels, the 1D kernels require significantly fewer parameters to be estimated. Our method develops a deep fully convolutional neural network that takes two input frames and estimates pairs of 1D kernels for all pixels simultaneously. Since our method is able to estimate kernels and synthesizes the whole video frame at once, it allows for the incorporation of perceptual loss to train the neural network to produce visually pleasing frames. This deep neural network is trained end-to-end using widely available video data without any human annotation. Both qualitative and quantitative experiments show that our method provides a practical solution to high-quality video frame interpolation.

Paper
Simon Niklaus, Long Mai, and Feng Liu. Video Frame Interpolation via Adaptive Separable Convolution
IEEE ICCV 2017. PDF  
Related Paper
Simon Niklaus, Long Mai, and Feng Liu. Video Frame Interpolation via Adaptive Convolution
IEEE CVPR 2017. PDF  Project Website(spotlight)
Demo Video 
Search engine friendly content
Acknowledgment
This work was supported by NSF IIS-1321119. This video uses materials under a Creative Common license or with the owner's permission, as detailed at the end.