A new passage ranking algorithm for video question answering

Yu Chieh Wu, Yue Shi Lee, Jie Chi Yang, Show Jane Yen

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

4 Scopus citations

Abstract

Developing a question answering (Q/A) system involves in integrating abundant linguistic resources such as syntactic parsers, named entity recognizers which are not only impose time cost but also unavailable in other languages. Ranking-based approaches take the advantage of both efficiency and multilingual portability but most of them bias to high frequent words. In this paper, we propose a new passage ranking algorithm for extending textQ/A toward videoQ/A based on searching lexical information in videos. This method takes both N-gram match and word density into account and finds the optimal match sequence using dynamic programming techniques. Besides, it is very efficient to handle real time tasks for online video question answering. We evaluated our method with 150 actual user's questions on the 45GB video collections. Nevertheless, four well-known but multilingual portable ranking approaches were adopted to compare. Experimental results show that our method outperforms the second best approach with relatively 25.64% MRR score.

Original languageEnglish
Title of host publicationAdvances in Image and Video Technology - First Pacific Rim Symposium, PSIVT 2006, Proceedings
PublisherSpringer Verlag
Pages563-572
Number of pages10
ISBN (Print)354068297X, 9783540682974
DOIs
StatePublished - 2006
Event1st Pacific Rim Symposium on Image and Video Technology, PSIVT 2006 - Hsinchu, Taiwan
Duration: 10 Dec 200613 Dec 2006

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume4319 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference1st Pacific Rim Symposium on Image and Video Technology, PSIVT 2006
Country/TerritoryTaiwan
CityHsinchu
Period10/12/0613/12/06

Fingerprint

Dive into the research topics of 'A new passage ranking algorithm for video question answering'. Together they form a unique fingerprint.

Cite this