Context-aware Synthesis for Video Frame Interpolation
Simon Niklaus and Feng Liu
Computer Science Department
Portland State University
Abstract

Video frame interpolation algorithms typically estimate optical flow or its variations and then use it to guide the synthesis of an intermediate frame between two consecutive original frames. To handle challenges like occlusion, bidirectional flow between the two input frames is often estimated and used to warp and blend the input frames. However, how to effectively blend the two warped frames still remains a challenging problem. This paper presents a context-aware synthesis approach that warps not only the input frames but also their pixel-wise contextual information and uses them to interpolate a high-quality intermediate frame. Specifically, we first use a pre-trained neural network to extract per-pixel contextual information for input frames. We then employ a state-of-the-art optical flow algorithm to estimate bidirectional flow between them and pre-warp both input frames and their context maps. Finally, unlike common approaches that blend the pre-warped frames, our method feeds them and their context maps to a video frame synthesis neural network to produce the interpolated frame in a context-aware fashion. Our neural network is fully convolutional and is trained end to end. Our experiments show that our method can handle challenging scenarios such as occlusion and large motion and outperforms representative state-of-the-art approaches.

Paper
Simon Niklaus and Feng Liu. Context-aware Synthesis for Video Frame Interpolation
IEEE CVPR 2018. PDF  (Rank 1st in the relevant Middlebury benchmark)
Related Paper
Simon Niklaus, Long Mai, and Feng Liu. Video Frame Interpolation via Adaptive Separable Convolution
IEEE ICCV 2017. PDF  Project Website Code


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
The example of the busy alley is used with permission from John Power. All the other examples were from the DAVIS Challenge, the Middlebury Computer Vision Benchmark, the Blender Foundation, the KITTI Benchmark, DVF (from UCF101), and RMIT3D..