Computer Vision

2014 Edition
| Editors: Katsushi Ikeuchi

Video Retrieval

  • Cees G. M. Snoek
  • Arnold W. M. Smeulders
Reference work entry



Video retrieval is the process of searching in video based on an analysis of its visual content.


The cause for the general video retrieval problem is the semantic gap: the lack of correspondence between the low-level features that machines extract from the visual signal and the high-level conceptual interpretations a human gives [1]. In order to bridge the gap, many retrieval solutions have been proposed in the past, e.g., by using text, speech, tags, or example images [2, 3]. But the most authentic and cognitive hardest is to type a concept from visual information and to retrieve the images carrying that concept [4].


The video retrieval method of choice in the field is rendered in Fig. 1. The first step is to extract from an image locally measured features, lots of them, ranging from 40 to 100,000. The features are invariant descriptors which cancel out accidental circumstances of the recording caused by...
This is a preview of subscription content, log in to check access.


  1. 1.
    Smeulders AWM, Worring M, Santini S, Gupta A, Jain R (2000) Content-based image retrieval at the end of the early years. IEEE Trans Pattern Anal Mach Intell 22(12):1349–1380CrossRefGoogle Scholar
  2. 2.
    Wactlar HD, Christel MG, Gong Y, Hauptmann AG (1999) Lessons learned from building a terabyte digital video library. IEEE Comput 32(2):66–73CrossRefGoogle Scholar
  3. 3.
    Smith JR, Chang SF (1997) Visually searching the web for content. IEEE MultiMed 4(3):12–20CrossRefGoogle Scholar
  4. 4.
    Snoek CGM, Worring M (2009) Concept-based video retrieval. Found Trends Inf Retr 4(2):215–322Google Scholar
  5. 5.
    Smeaton AF, Over P, Kraaij W (2006) Evaluation campaigns and TRECVid. In: Proceedings of the ACM SIGMM international workshop on multimedia information retrieval. ACM, New York, pp 321–330Google Scholar
  6. 6.
    Naphade MR, Smith JR, Tešić J, Chang SF, Hsu W, Kennedy LS, Hauptmann AG, Curtis J (2006) Large-scale concept ontology for mul- timedia. IEEE MultiMed 13(3): 86–91CrossRefGoogle Scholar
  7. 7.
    Snoek CGM, Smeulders AWM (2010) Visual-concept search solved? IEEE Comput 43(6):76–78CrossRefGoogle Scholar
  8. 8.
    Hauptmann AG, Yan R, Lin WH, Christel MG, Wactlar H (2007) Can high-level concepts fill the semantic gap in video retrieval? A case study with broadcast news. IEEE Trans Multimed 9(5):958–966CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2014

Authors and Affiliations

  • Cees G. M. Snoek
    • 1
    • 3
  • Arnold W. M. Smeulders
    • 2
    • 3
  1. 1.University of AmsterdamAmsterdamThe Netherlands
  2. 2.Centre for Mathematics and Computer Science (CWI), University of AmsterdamAmsterdamThe Netherlands
  3. 3.Intelligent Systems Lab Amsterdam, Informatics Institute University of AmsterdamAmsterdamThe Netherlands