THIS PAGE IS NO LONGER UPDATED
You will be redirected to the new site in 5 secs...
Welcome to the Image & Signal Processing group!








Human Vision, Image Models and Image Processing at the Universitat de València


Image Statistics

Vision Science

Image Processing

Video Processing

Color Science

Software

Publications

People
projects
Projects

Links

(Clicking on the VISTA icons will take you back to this page)



Overview
 

 

 


         


OVERVIEW: VISION AND IMAGE STATISTICS

The Barlow's hypothesis
The structure of natural objects gives rise to images with very special statistical features. For an observer with knowledge of this structure (or statistical features) the images in the spatial domain are highly redundant because each pixel can be predicted from its spatio temporal neighbors. Horace Barlow applied Information Theory to Vision Science and suggested that, any biological or artificial vision system should organize its sensors (should represent the images) in such a way that their responses are statistically independent and are useful to solve classification problems [1].

This simple hypothesis, which relates Information Theory, Pattern Recognition and Vision Science in a very fundamental way [1,2,3], is the inspiring idea of our work.

Our research interests
Taking into account the above relations, our general goal is to achieve a deeper understanding of natural image statistics because from this knowledge it should be possible to explain the behavior of the visual cortex and propose new alternatives in a number of applications in image processing and computer vision in which the basic problem is the choice of an appropriate signal representation [4]. The range of basic and applied topics in which we are currently working include:


Nice references (even though they are not from us!)

[1]  H.B. Barlow. Redundancy Reduction Revisited.
Network: Computation in Neural Systems. Vol. 12. pp. 241- 253 (2001) 
(Abstract and PDF file from Barlow's site)
(PDF file from our site)
[2]   E.P. Simoncelli. Vision and the Statistics of the Visual Environment.
Current Opinion in Neurobiology, Vol. 13. (2003) 
(Abstract and PDF file from Simoncelli's site)
(PDF file from our site)
[3]  E.P. Simoncelli and B.O. Olshausen. Natural image statistics and neural representation
Annual Review of Neuroscience, Vol. 24. pp.1193-1216 (2001) 
(Abstract and PDF file from Simoncelli's site)
(PDF file from our site)
[4]  E.P. Simoncelli. Statistical Models for Images: Compression, Restoration and Synthesis. 
31st Asilomar Conference on Signals, Systems and Computers, Pacific Grove, CA. November 2-5, (1997)
(Abstract and PDF file from Simoncelli's site)
(PDF file from our site)
 


 


 


VISTA is a larger project in colaboration with the Granada University. To see our global web page, click HERE