Access Restriction

Author Yadong Mu ♦ Wei Liu ♦ Shuicheng Yan
Sponsorship IEEE Circuits and Systems Society
Source IEEE Xplore Digital Library
Content type Text
Publisher Institute of Electrical and Electronics Engineers, Inc. (IEEE)
File Format PDF
Copyright Year ©1991
Language English
Subject Domain (in DDC) Computer science, information & general works ♦ Data processing & computer science ♦ Natural sciences & mathematics ♦ Physics ♦ Electricity & electronics
Subject Keyword Cameras ♦ Visualization ♦ Image restoration ♦ Robustness ♦ Rain ♦ Optical imaging ♦ Estimation ♦ sub-pixel alignment ♦ Video de-fencing ♦ weighted truncated optical flow ♦ weighted truncated optical flow ♦ Subpixel alignment ♦ video de-fencing
Abstract This paper describes and provides an initial solution to a novel video editing task, i.e., video de-fencing. It targets automatic restoration of the video clips that are corrupted by fence-like occlusions during capture. Our key observation lies in the visual parallax between fences and background scenes, which is caused by the fact that the former are typically closer to the camera. Unlike in traditional image inpainting, fence-occluded pixels in the videos tend to appear later in the temporal dimension and are therefore recoverable via optimized pixel selection from relevant frames. To eventually produce fence-free videos, major challenges include cross-frame subpixel image alignment under diverse scene depth, and correct pixel selection that is robust to dominating fence pixels. Several novel tools are developed in this paper, including soft fence detection, weighted truncated optical flow method, and robust temporal median filter. The proposed algorithm is validated on several real-world video clips with fences.
Description Author affiliation :: AT&T Labs. - Res., Middletown, NJ, USA
Author affiliation :: IBM T. J. Watson Res. Center, Yorktown Heights, NY, USA
Author affiliation :: Dept. of Electr. & Comput. Eng., Nat. Univ. of Singapore, Singapore, Singapore
ISSN 10518215
Education Level UG and PG
Learning Resource Type Article
Publisher Date 2014-01-01
Publisher Place U.S.A.
Rights Holder Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Volume Number 24
Issue Number 7
Size (in Bytes) 14.59 MB
Page Count 11
Starting Page 1111
Ending Page 1121

Source: IEEE Xplore Digital Library