Query on demand video browsing

Abstract

This paper describes a novel method for browsing a large collection of news video by linking various forms of related video fragments together as threads. Each thread contains a sequence of shots with high feature-based similarity. Two interfaces are designed which use threads as the basis for browsing. One interface shows a minimal set of threads, and the other as many as possible. Both interfaces are evaluated in the TRECVID interactive retrieval task, where they ranked among the best interactive retrieval systems currently available. The results indicate that the use of threads in interactive video search is very beneficial. We have found that in general the query result and the timeline are the most important threads. However, having several additional threads allow a user to find unique results which cannot easily be found by using query results and time alone.

Video browsing interfaces and applications: a review

“…de Rooij et al [105] and Snoek et al [106] introduced the notion of video threads, where a thread is a sequence of feature-based similar shots from several videos in some specific order. They differentiate between (1) visual threads having visual similarity, (2) textual threads having similar textual annotations, (3) semantic threads having semantically equivalent shots, (4) time threads having temporal similarity, (5) query result threads having similarity according to a query, and (6) history threads consisting of shots the user has already visited.…”

Video Browsing Using Interactive Navigation Summaries

“…Goeau et al [9] proposed the so called Table of Video Contents (TOVC) visualization for browsing story-based video content like news, interviews, or sports summaries. Rooij et al [6] introduced the concept of video threads, where a video thread is a sequence of feature-based similar shots from several videos in some specific order. A visualization of video threads for the purpose of explorative browsing has been developed as a CrossBrowser and a RotorBrowser.…”

Content‐based video retrieval: Three example systems from TRECVid

“…For TRECVid 2007, the focus was specifically on consolidation of proven interface components from previous TRECVid editions into a novel browsing environment (Snoek et al, 2007a) The notion of threads was introduced in the ForkBrowser in order to browse through a video data set in multiple directions. A thread is a linked sequence of shots in a specified order, based upon an aspect of their content (de Rooij et al, 2007) including static threads which are pre-computed, and dynamic threads which are generated on demand. The content of a thread is based on a form of similarity between shots in the data set.…”

“…For topics that require a combination of threads, the RotorBrowser was introduced in 2006 (Snoek et al, 2006;de Rooij et al, 2007). This allows a user to integrate query results with time, visual similarity, semantic similarity and various other shot-based similarity metrics.…”