Video De-fencing

TL;DR

This paper introduces a novel video de-fencing method that leverages temporal information and scene depth cues to automatically remove fence-like occlusions from videos, improving visual quality.

Contribution

It proposes new techniques such as soft fence detection, weighted optical flow, and robust temporal median filtering for effective fence removal in videos.

Findings

Successfully removes fences from real-world videos

Achieves accurate cross-frame pixel alignment

Demonstrates robustness to diverse scene depths

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 sub-pixel 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.