An efficient color representation for image retrieval

By Yining Deng, Bangalore S. Manjunath, C. Kenney, M. Moore, H. Shin. IEEE Transactions on Image Processing, 2001

360 citations recorded by the source corpus.

A compact color descriptor and an efficient indexing method for this descriptor are presented. The target application is similarity retrieval in large image databases using color. Colors in a given region are clustered into a small number of representative colors. The feature descriptor consists of the representative colors and their percentages in the region. A similarity measure similar to the quadratic color histogram distance measure is defined for this descriptor. The representative colors can be indexed in the three-dimensional (3-D) color space thus avoiding the high-dimensional indexing problems associated with the traditional color histogram. For similarity retrieval, each representative color in the query image or region is used independently to find regions containing that color. The matches from all of the query colors are then combined to obtain the final retrievals. An efficient indexing scheme for fast retrieval is presented. Experimental results show that this compact descriptor is effective and compares favorably with the traditional color histogram in terms of overall computational complexity.

View Paper: https://www.semanticscholar.org/paper/6e27a4e623e5b14f4b88907d79df153daf3716ab

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Color and texture descriptors (2001, methodology): “In order to compute this descriptor, the colors present in a given image or region are first clustered (see [4] and [7] for more details).”
Image retrieval: Ideas, influences, and trends of the new age (2008, methodology): “In [Deng et al. 2001; Wang et al. 2001], it is argued that region-based signature is more efficient computationally for retrieval, and it also gets around drawbacks associated with earlier propositions such as dimension reduction and color moment descriptors.”
Content-based image retrieval: approaches and trends of the new age (2005, unknown): “the regions is proposed in [25], and shown to be more computationally efficient in similarity based retrieval than traditional color histograms.”
Dissimilarity-based people re-identification and search for intelligent video surveillance (2013, methodology): “Given an image, the DCD algorithm first finds the K dominant colours [36], via k-means clustering of all the colour triplets in the image.”
OPTIMOL: Automatic Online Picture Collection via Incremental Model Learning (2007, methodology): “Image Retrieval from the Web Content-based image retrieval (CBIR) (Zhou and Huang 2002; Deng 2001; Carson et al. 1999; Li et al. 2000; Chen et al. 2003; Jain and Vailaya 1996; Barnard and Forsyth 2001; Barnard et al. 2003; Jeon et al. 2003) has been long an active ﬁeld of research.”