Skip to main navigation Skip to search Skip to main content

Geodesic distance histogram feature for video segmentation

  • Stony Brook University
  • Harvard University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

10 Scopus citations

Abstract

This paper proposes a geodesic-distance-based feature that encodes global information for improved video segmentation algorithms. The feature is a joint histogram of intensity and geodesic distances, where the geodesic distances are computed as the shortest paths between superpixels via their boundaries. We also incorporate adaptive voting weights and spatial pyramid configurations to include spatial information into the geodesic histogram feature and show that this further improves results. The feature is generic and can be used as part of various algorithms. In experiments, we test the geodesic histogram feature by incorporating it into two existing video segmentation frameworks. This leads to significantly better performance in 3D video segmentation benchmarks on two datasets.

Original languageEnglish
Title of host publicationComputer Vision - ACCV 2016 - 13th Asian Conference on Computer Vision, Revised Selected Papers
EditorsYoichi Sato, Ko Nishino, Vincent Lepetit, Shang-Hong Lai
PublisherSpringer Verlag
Pages275-290
Number of pages16
ISBN (Print)9783319541808
DOIs
StatePublished - 2017
Event13th Asian Conference on Computer Vision, ACCV 2016 - Taipei, Taiwan, Province of China
Duration: Nov 20 2016Nov 24 2016

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10111 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference13th Asian Conference on Computer Vision, ACCV 2016
Country/TerritoryTaiwan, Province of China
City Taipei
Period11/20/1611/24/16

Fingerprint

Dive into the research topics of 'Geodesic distance histogram feature for video segmentation'. Together they form a unique fingerprint.

Cite this