Open AccessA PatchMatch-based Approachfor Matte Propagation in Videos
Author(s) -
Marcos Henrique Backes,
Manuel M. Oliveira Neto
Publication year - 2020
Language(s) - English
Resource type - Conference proceedings
DOI - 10.5753/ctd.2020.11367
Subject(s) - computer science , artificial intelligence , computer vision , enhanced data rates for gsm evolution , computer graphics (images) , video editing , sequence (biology) , video quality , metric (unit) , operations management , genetics , economics , biology
This thesis presents a temporally-coherent matte-propagation method for videos based on PatchMatch and edge-aware filtering. Given an input video and trimaps for a few frames, including the first and last, our approach generates alpha mattes for all frames of the video sequence. We also present a user scribble-based interface for video matting that takes advantage of the efficiency of our method to interactively refine the matte results. We perform quantitative comparisons against the state-of-the-art sparse-input video matting techniques and show that our method produces significantly better results according to three different metrics. We also perform qualitative comparisons against the state-of-the-art dense-input video matting techniques and show that ours produces similar quality results while requiring less than 7% of their user input.