Natural images are highly structured in their spatial configuration. Where one would expect a different spatial distribution for every image, as each image has a different spatial layout, we show that the spatial statistics of recorded images can be explained by a single process of sequential fragmentation. The observation by a resolution limited sensory system turns out to have a profound influence on the observed statistics of natural images. The power-law and normal distribution represent the extreme cases of sequential fragmentation. Between these two extremes, spatial detail statistics deform from power-law to normal through the Weibull type distribution as receptive field size increases relative to image detail size.
@InProceedings{GeusebroekICCV2003,
author = "Geusebroek, J. M. and Smeulders, A. W. M.",
title = "Fragmentation in the Vision of Scenes",
booktitle = "IEEE International Conference on Computer Vision",
volume = "1",
pages = "130--135",
year = "2003",
url = "https://ivi.fnwi.uva.nl/isis/publications/2003/GeusebroekICCV2003",
pdf = "https://ivi.fnwi.uva.nl/isis/publications/2003/GeusebroekICCV2003/GeusebroekICCV2003.pdf",
has_image = 1
}