Content-Based Image Retrieval (CBIR)

1271 Words3 Pages

The biggest issue when it comes to creating a system for image retrieval based on image queries is the semantic gap. The semantic gap is defined as “the lack of coincidence between the information that one can extract from the visual data and the interpretation that the same data have for a user in a given situation” (Smeulders). What a system can discern from an image can be completely different to what a human user can discern. It can also be the case that two human users assign different descriptions for the same image, adding to the complexity of the problem.

I first got interested in this topic when I first heard of content-based image retrieval (CBIR) about a year ago. A year ago, CBIR systems were still developing and maturing. Most results returned by these systems when presented with an image query back then were images that were almost identical to the query. Today, systems are able to identify the main focus of objects in pictures and return a more varied assortment of results. This in particular is what interest me the most, how such systems came to be more accurate in giving more appropriate results to the end user.

The main component to CBIR systems is Vision Science. Vision Science is the “study of how humans see and interpret the light that lands on the sensor known as the retina” (Palmer). There are five key research points within vision science that relate to CBIR (Marques). First is attention, the concern with how the human visual system (HVS) prioritizes and selects what region of a scene it attends to. Second, perception, the interpretation of sampled visual information. Memory, access to past knowledge, rules, and intuition, as well as the recording of the current imagery. Contextual Effects, the environ...

... middle of paper ...

...ww.sdsc.edu/~gupta/publications/PAMI-01-review.pdf

http://neurobio.drexelmed.edu/Rybakweb/vr.pdf

S. Palmer. Vision Science: Photons to Phenomenology. MIT Press, Cambridge, MA, 1999.

O. Marques, L. M. Mayron, G. B. Borba, and H. R. Gamba. On the potential of incorporating knowledge of human visual attention into CBIR systems. In Special Session on Perceptual Visual Processing, IEEE International Conference on Multimedia & Expo (ICME 2006), Toronto, Canada, July 2006

D. Noton and L. Stark. Scanpaths in eye movements during pattern perception. Science, 171:308-311, Jan. 1971

L. Itti, C. Koch, and E. Niebur. A model of saliency-based visual attention for rapid scene analysis. IEEE Trans. on PAMI, 20(11):1245-1259, Nov 1998

D. Jobson, Z. Rahman, and G. Woodell. Properties and performance of a center/surround retinex. IP, 6(3):451-462, March 1997

Open Document