# Real-time Deep Video Deinterlacing Source: [Real-time Deep Video Deinterlacing]() Collection index: [llms.txt]() Data updated: 2026-01-04T02:50:22.754Z Authors: Haichao Zhu, Xueting Liu, Xiangyu Mao, Tien-Tsin Wong Published: 2017-08-01T07:23:53.000Z ## Abstract Interlacing is a widely used technique, for television broadcast and video recording, to double the perceived frame rate without increasing the bandwidth. But it presents annoying visual artifacts, such as flickering and silhouette "serration," during the playback. Existing state-of-the-art deinterlacing methods either ignore the temporal information to provide real-time performance but lower visual quality, or estimate the motion for better deinterlacing but with a trade-off of higher computational cost. In this paper, we present the first and novel deep convolutional neural networks (DCNNs) based method to deinterlace with high visual quality and real-time performance. Unlike existing models for super-resolution problems which relies on the translation-invariant assumption, our proposed DCNN model utilizes the temporal information from both the odd and even half frames to reconstruct only the missing scanlines, and retains the given odd and even scanlines for producing the full deinterlaced frames. By further introducing a layer-sharable architecture, our system can achieve real-time performance on a single GPU. Experiments shows that our method outperforms all existing methods, in terms of reconstruction accuracy and computational performance. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Video Deinterlacing on Video Deinterlacing Sequences (test)](): Video Deinterlacing; Video Deinterlacing Sequences (test); Average Latency (s): 0.0137. - [Video Deinterlacing on Taxi (test)](): Video Deinterlacing; Taxi (test); PSNR: 38.15. - [Video Deinterlacing on Basketball (test)](): Video Deinterlacing; Basketball (test); PSNR: 36.55. - [Video Deinterlacing on Jumping (test)](): Video Deinterlacing; Jumping (test); PSNR: 39.75. - [Video Deinterlacing on Tide (test)](): Video Deinterlacing; Tide (test); PSNR: 35.37. - [Video Deinterlacing on Girl (test)](): Video Deinterlacing; Girl (test); PSNR: 35.44. - [Video Deinterlacing on Roof (test)](): Video Deinterlacing; Roof (test); PSNR: 35.44. - [Video Deinterlacing, Super-Resolution, and Frame Interpolation on Youku2K 540i(25P)-to-1080p(50P) (test)](): Video Deinterlacing, Super-Resolution, and Frame Interpolation; Youku2K 540i(25P)-to-1080p(50P) (test); PSNR: 40.3. - [Video Deinterlacing, Super-Resolution, and Frame Interpolation on SJTU 1080i(25P)-to-4K (test)](): Video Deinterlacing, Super-Resolution, and Frame Interpolation; SJTU 1080i(25P)-to-4K(50P) (test); PSNR: 39.04. - [Video Deinterlacing and Super-Resolution on Youku2K 540i-to-1080P (test)](): Video Deinterlacing and Super-Resolution; Youku2K 540i-to-1080P (test); PSNR: 40.29. - [Video Deinterlacing and Super-Resolution on SJTU 1080i-to-4K (test)](): Video Deinterlacing and Super-Resolution; SJTU 1080i-to-4K (test); PSNR: 39.05. - [Deinterlacing on YOUKU 2K synthetic (test)](): Deinterlacing; YOUKU 2K synthetic (test); PSNR: 29.16. - [Deinterlacing on SJTU 4K synthetic (test)](): Deinterlacing; SJTU 4K synthetic (test); PSNR: 29.3. - [Deinterlacing on Vimeo 90K synthetic (test)](): Deinterlacing; Vimeo 90K synthetic (test); PSNR: 29.87. ## Original sources and implementation [Original paper]() [PDF]()