標題: DEEP REINFORCEMENT LEARNING FOR VIDEO PREDICTION
作者: Ho, Yung-Han
Cho, Chuan-Yuan
Peng, Wen-Hsiao
資訊工程學系
Department of Computer Science
關鍵字: Reinforcement learning;deep video prediction
公開日期: 1-一月-2019
摘要: This paper introduces a hybrid video prediction scheme that combines the classic parametric overlapped block motion compensation (POBMC) technique with neural networks. Most learning-based video prediction methods rely on a black-box-like model for either direct generation of future video frames or estimation of a dense motion field. The model complexity often increases drastically with frame resolution. Departing from pure black-box approaches, this paper leverages the theoretically-grounded POBMC in a reinforcement learning framework to estimate a sparse motion field for future frame warping. Two neural networks are trained to identify critical points in the motion field for motion estimation. We train our model on 10k unlabeled frames in KITTI dataset and achieve the state-of-the-art SSIM score of 0.923 on CaltechPed and an average SSIM scroe of 0.856 on Common Intermediate Format (CIF) standard sequences.
URI: http://hdl.handle.net/11536/154040
ISBN: 978-1-5386-6249-6
ISSN: 1522-4880
期刊: 2019 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP)
起始頁: 604
結束頁: 608
顯示於類別:會議論文