LOOKING-AHEAD: NEURAL FUTURE VIDEO FRAME PREDICTION

被引：0

作者：

Zhang, Changxu ^{[1
]}

Chen, Tong ^{[1
]}

Liu, Haojie ^{[1
]}

Shen, Qiu ^{[1
]}

Ma, Zhan ^{[1
]}

机构：

[1] Nanjing Univ, Vis Lab, Nanjing, Peoples R China

来源：

2019 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP) | 2019年

基金：

中国国家自然科学基金;

关键词：

Optical flow; inpainting; deep neural networks; video frame prediction;

D O I：

10.1109/icip.2019.8803151

中图分类号：

TB8 [摄影技术];

学科分类号：

0804 ;

摘要：

We have developed a Looking-Ahead system to facilitate the future video frame prediction via deep learning, which is of practical value in the domain like autonomous driving etc. The overall problem is decomposed into cascaded optical flow prediction and subsequent predictive frame post-processing for quality refinement. A pyramid flow calculation across existing frames is used to efficiently infer the motion of target frame; while a universal inpainting network is applied to restore those motion-induced occluded pixels. Compared with those published methods, our Looking-Ahead offers the state-of-the-art performance measured objectively with better Peak-Signal-to-Noise Ratio (PSNR) and Structural Similarity (SSIM), and more appealing reconstructions.

引用

页码：1975 / 1979

页数：5