Yet Another Computer Vision Index To Datasets (YACVID) - Details

Stand: 2017-04-28 000000m 18:01:04 - Overview

ok542
Attribute Current Content New
Name (Institute + Shorttitle)Domain-specific Personal Videos Highlight Dataset 
Description (include details on usage, files and paper references)The domain-specific personal videos highlight dataset from the paper [1] describes a fully automatic method to train domain-specific highlight ranker for raw personal videos. Given a keyword such as surfing, the proposed method automatically harvests YouTube for both raw and edited pairs of personal videos for training a domain-specific highlight ranker. Experimental results demonstrate state-of-the-art highlight detection ability on surfing, skating, skiing, gymnastics, parkour, and dog related domains.

[1] M. Sun, A. Farhadi, and S. Seitz. Ranking Domain-specific Highlights by Analyzing Edited Videos. ECCV, 2014 
URL Linkhttps://github.com/aliensunmin/DomainSpecificHighlight 
Files (#)
References (SKIPPED)
Category (SKIPPED) 
Tags (single words, spaced)video summarization saliency wearable human action recognition domain 
Last Changed2017-04-28 
Turing (2.12+3.25=?) :-)