This paper presents a new deformable convolution-based video frame interpolation (VFI) method, using a coarse to fine 3D CNN to enhance the multi-flow prediction. This model first extracts spatio-temporal features at multiple scales using a 3D CNN, and estimates multi-flows using these features in a coarse-to-fine manner. The estimated multi-flows are then used to warp the original input frames as well as context maps, and the warped results are fused by a synthesis network to produce the final output. This VFI approach has been fully evaluated against 12 state-of-the-art VFI methods on three commonly used test databases. The results evidently show the effectiveness of the proposed method, which offers superior interpolation performance over other state of the art algorithms, with PSNR gains up to 0.19dB.

翻译：本文介绍了一种新的变形变形的基于变形的视频框架内插法(VFI)方法,使用粗略到微小的3DCNN, 以加强多流量预测。 这个模型首先用 3D CNN 提取多个比例尺的时空特征, 并以粗略到松散的方式估算使用这些特征的多流量。 然后, 估计的多流量用于扭曲原始输入框架和上下文图, 扭曲的结果由合成网络结合, 以生成最终输出。 这个 VFI 方法已经根据三个常用测试数据库的12种最先进的VFI方法进行了充分评价。 结果表明, 所拟议的方法的功效是优于艺术算法的其他状态, PSNR 收益高达 0. 19dB 。