Semantic search

ABSTRACT

We review the state of the art in content-based exploration of large sets of image and video data. The key to any current approach is features to which still many new additions have been made in the last five years). We discuss image features on a global, regional, and key point level employing stochastic textures derived from natural image statistics and Gaussian color invariant feature sets [Geusebroek 2006]. Features evolving from analysis moving images are still under the largest evolution. In this topic we will discuss several advances towards robust and long-term tracking [Nguyen 2005, Nguyen 2006, Burghouts 2006.

Then we move on to review the state of the art in retrieval systems as it has evolved since "the end of the early years" in 2000. We do so on the basis of the components and evaluation of our video search engine. The MediaMill system performed well in the TRECvid competition in the years 2004 to 2006 [Snoek 2006] and provide a perspective the future of search engines.

REFERENCES

• 1: Hieu Tat Nguyen and Arnold W. M. Smeulders: Multiple target tracking by incremental probabilistic PCA, IEEE trans PAMI 29-1, 52--64, 2007.
• 2: Hieu T. Nguyen , Arnold W. Smeulders, Robust Tracking Using Foreground-Background Texture Discrimination, International Journal of Computer Vision, v.69 n.3, p.277-293, September 2006 [doi>10.1007/s11263-006-7067-x]
• 3: Jan C. van Gemert, Jan-Mark Geusebroek, Cor J. Veenman, Cees G. M. Snoek Arnold W. M. Smeulders: Robust Scene Categorization by Learning Image Statistics in Context, SLAM06.
• 4: Cees G. M. Snoek , Marcel Worring , Jan-Mark Geusebroek , Dennis C. Koelma , Frank J. Seinstra , Arnold W. M. Smeulders, The Semantic Pathfinder: Using an Authoring Metaphor for Generic Multimedia Indexing, IEEE Transactions on Pattern Analysis and Machine Intelligence, v.28 n.10, p.1678-1689, October 2006 [doi>10.1109/TPAMI.2006.212]
• 5: Gert Jan Burghouts and Jan Mark Geusebroek: Quasi-periodic spatiotemporal filtering. IEEE trans IP, 15-6, 1572--1582, 2006.

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.4 IMAGE PROCESSING AND COMPUTER VISION
      I.4.7 Feature Measurement
          Subjects: Feature representation

General Terms:
Algorithms, Human Factors