Can High-Level Concepts Fill the Semantic Gap in Video Retrieval? A Case Study With Broadcast News
- 23 July 2007
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 9 (5), 958-966
- https://doi.org/10.1109/tmm.2007.900150
Abstract
A number of researchers have been building high-level semantic concept detectors such as outdoors, face, building, to help with semantic video retrieval. Our goal is to examine how many concepts would be needed, and how they should be selected and used. Simulating performance of video retrieval under different assumptions of concept detection accuracy, we find that good retrieval can be achieved even when detection accuracy is low, if sufficiently many concepts are combined. We also derive suggestions regarding the types of concepts that would be most helpful for a large concept lexicon. Since our user study finds that people cannot predict which concepts will help their query, we also suggest ways to find the best concepts to use. Ultimately, this paper concludes that "concept-based" video retrieval with fewer than 5000 concepts, detected with a minimal accuracy of 10% mean average precision is likely to provide high accuracy results in broadcast news retrieval.Keywords
This publication has 17 references indexed in Scilit:
- The Semantic Pathfinder: Using an Authoring Metaphor for Generic Multimedia IndexingIEEE Transactions on Pattern Analysis and Machine Intelligence, 2006
- Large-Scale Concept Ontology for MultimediaIEEE MultiMedia, 2006
- Exploring Automatic Query Refinement for Text-Based Video RetrievalPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2006
- Video Retrieval Using High Level Features: Exploiting Query Matching and Confidence-Based WeightingLecture Notes in Computer Science, 2006
- Face recognitionACM Computing Surveys, 2003
- TRECVID: Benchmarking the Effectiveness of Information Retrieval Tasks on Digital VideoLecture Notes in Computer Science, 2003
- Automatic image annotation and retrieval using cross-media relevance modelsPublished by Association for Computing Machinery (ACM) ,2003
- Content-based image retrieval at the end of the early yearsIeee Transactions On Pattern Analysis and Machine Intelligence, 2000
- Concept based query expansionPublished by Association for Computing Machinery (ACM) ,1993
- Digital Signal Processing Methods for Ultrasonic Backscattered Waves in Composite MaterialsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1986