Learning to Perceive Transparency from the Statistics of Natural Scenes

NeurIPS 2002 pp. 1271-1278

/neurips/2002/levin2002neurips-learning/

Abstract

Certain simple images are known to trigger a percept of trans- parency: the input image I is perceived as the sum of two images I(x; y) = I1(x; y) + I2(x; y). This percept is puzzling. First, why do we choose the \more complicated" description with two images rather than the \simpler" explanation I(x; y) = I1(x; y) + 0 ? Sec- ond, given the inﬂnite number of ways to express I as a sum of two images, how do we compute the \best" decomposition ? Here we suggest that transparency is the rational percept of a sys- tem that is adapted to the statistics of natural scenes. We present a probabilistic model of images based on the qualitative statistics of derivative ﬂlters and \corner detectors" in natural scenes and use this model to ﬂnd the most probable decomposition of a novel image. The optimization is performed using loopy belief propa- gation. We show that our model computes perceptually \correct" decompositions on synthetic images and discuss its application to real images.

PDF NeurIPS Semantic Scholar

Cite

Text

Levin et al. "Learning to Perceive Transparency from the Statistics of Natural Scenes." Neural Information Processing Systems, 2002.

Markdown

[Levin et al. "Learning to Perceive Transparency from the Statistics of Natural Scenes." Neural Information Processing Systems, 2002.](https://mlanthology.org/neurips/2002/levin2002neurips-learning/)

BibTeX

@inproceedings{levin2002neurips-learning,
  title     = {{Learning to Perceive Transparency from the Statistics of Natural Scenes}},
  author    = {Levin, Anat and Zomet, Assaf and Weiss, Yair},
  booktitle = {Neural Information Processing Systems},
  year      = {2002},
  pages     = {1271-1278},
  url       = {https://mlanthology.org/neurips/2002/levin2002neurips-learning/}
}