Detection of Binocular Disparities
Overview
Authors
Affiliations
A stereo correspondence algorithm designed to perform matching on figurally similar images (arising in normal human binocular vision) is described. It is based on the observation that the operational principles underlying biological stereo disparity detection seem to be extremely general and few in number instead of an extended set of specific "constraints". We identify one general characteristic of objects in the three dimensional world and use it to formulate a simple noniterative, parallel and local algorithm that successfully detects disparities generated by opaque as well as transparent surfaces.
Zhang Y, Li X, Samonds J, Lee T Vision Res. 2015; 120:121-31.
PMID: 26712581 PMC: 4783228. DOI: 10.1016/j.visres.2015.12.002.
Solving stereo transparency with an extended coarse-to-fine disparity energy model.
Li Z, Qian N Neural Comput. 2015; 27(5):1058-82.
PMID: 25710090 PMC: 6223020. DOI: 10.1162/NECO_a_00722.
Samonds J, Potetz B, Lee T J Neurosci. 2009; 29(50):15780-95.
PMID: 20016094 PMC: 5788183. DOI: 10.1523/JNEUROSCI.2305-09.2009.
Limits of stereopsis explained by local cross-correlation.
Filippini H, Banks M J Vis. 2009; 9(1):8.1-18.
PMID: 19271878 PMC: 2940423. DOI: 10.1167/9.1.8.
Early computational processing in binocular vision and depth perception.
Read J Prog Biophys Mol Biol. 2004; 87(1):77-108.
PMID: 15471592 PMC: 1414095. DOI: 10.1016/j.pbiomolbio.2004.06.005.