Shape perception of water in photo-realistic 3D images. Sultana, A. & Allison, R. S. In 12th International Conference on Light and Color in Nature. 2016. abstract bibtex Light plays an extremely important role in the perception of transparency, depth and shape of water. In computer graphics, an important question is: How does perception of transparent objects (including 3D shape reconstruction) depend on the fidelity of the rendering. In this paper, I present a theoretical method to recover the surface shape of water under day light settings based on human visual stereoscopic view. Flat transparent objects with arbitrary depth should pass all the incoming light and we should obtain a clear view of the background through them. However, when the surface is not flat, the image of the background or the underlying surface gets distorted because the chromatic light passing through different regions of water surface experience distortion and absorption that varies with wavelength. The reflection and refraction angles of light hitting the surface of a material depend on the direction of the light, its spectral composition, the medium and its surface shape. In this study, we evaluate and improve the cues available for perceiving shape of refractive objects by exploring the relationship in a 3D view between (a) Reflective highlights on the water surface that depends largely on the lightning conditions and refractive features (with known index of refraction) seen through the surface of the medium, (b) viewing and perceiving conditions (stereoscopic and/or non stereoscopic) and (c) textures and shading that provide cues to distortions. Analysis to date predicts that humans should have information to identify and reconstruct shape of an object in refractive stereo given that: i. Object is transparent and visible ii. Scene redirects incoming light just once and index of refraction is known. iii. Surface is both optically smooth and textured iv. At least two viewpoints should be available to obtain one 3D point on its light path at viewing surface. Psychophysical experiments that we are undertaking will either confirm or falsify human perceptual capability to identify and reconstruct shape of object under the above conditions.
@incollection{Sultana:2016aa,
abstract = {Light plays an extremely important role in the perception of transparency, depth and shape of water. In computer graphics, an important question is: How does perception of transparent objects (including 3D shape reconstruction) depend on the fidelity of the rendering. In this paper, I present a theoretical method to recover the surface shape of water under day light settings based on human visual stereoscopic view. Flat transparent objects with arbitrary depth should pass all the incoming light and we should obtain a clear view of the background through them. However, when the surface is not flat, the image of the background or the underlying surface gets distorted because the chromatic light passing through different regions of water surface experience distortion and absorption that varies with wavelength. The reflection and refraction angles of light hitting the surface of a material depend on the direction of the light, its spectral composition, the medium and its surface shape. In this study, we evaluate and improve the cues available for perceiving shape of refractive objects by exploring the relationship in a 3D view between (a) Reflective highlights on the water surface that depends largely on the lightning conditions and refractive features (with known index of refraction) seen through the surface of the medium, (b) viewing and perceiving conditions (stereoscopic and/or non stereoscopic) and (c) textures and shading that provide cues to distortions.
Analysis to date predicts that humans should have information to identify and reconstruct shape of an object in refractive stereo given that:
i. Object is transparent and visible ii. Scene redirects incoming light just once and index of refraction is known.
iii. Surface is both optically smooth and textured iv. At least two viewpoints should be available to obtain one 3D point on its light path at viewing surface.
Psychophysical experiments that we are undertaking will either confirm or falsify human perceptual capability to identify and reconstruct shape of object under the above conditions.},
annote = {31 May - 3 June 2016 at the University of Granada, Spain},
author = {Sultana, A. and Allison, R. S.},
booktitle = {12th International Conference on Light and Color in Nature},
date-added = {2016-12-04 21:56:23 +0000},
date-modified = {2016-12-04 21:57:13 +0000},
keywords = {Stereopsis},
title = {Shape perception of water in photo-realistic 3D images},
year = {2016}}
Downloads: 0
{"_id":"QJb3JZMpDNwXuZ4Zk","bibbaseid":"sultana-allison-shapeperceptionofwaterinphotorealistic3dimages-2016","downloads":0,"creationDate":"2018-11-16T15:32:55.644Z","title":"Shape perception of water in photo-realistic 3D images","author_short":["Sultana, A.","Allison, R. S."],"year":2016,"bibtype":"incollection","biburl":"https://bibbase.org/network/files/ibWG96BS4w7ibooE9","bibdata":{"bibtype":"incollection","type":"incollection","abstract":"Light plays an extremely important role in the perception of transparency, depth and shape of water. In computer graphics, an important question is: How does perception of transparent objects (including 3D shape reconstruction) depend on the fidelity of the rendering. In this paper, I present a theoretical method to recover the surface shape of water under day light settings based on human visual stereoscopic view. Flat transparent objects with arbitrary depth should pass all the incoming light and we should obtain a clear view of the background through them. However, when the surface is not flat, the image of the background or the underlying surface gets distorted because the chromatic light passing through different regions of water surface experience distortion and absorption that varies with wavelength. The reflection and refraction angles of light hitting the surface of a material depend on the direction of the light, its spectral composition, the medium and its surface shape. In this study, we evaluate and improve the cues available for perceiving shape of refractive objects by exploring the relationship in a 3D view between (a) Reflective highlights on the water surface that depends largely on the lightning conditions and refractive features (with known index of refraction) seen through the surface of the medium, (b) viewing and perceiving conditions (stereoscopic and/or non stereoscopic) and (c) textures and shading that provide cues to distortions. Analysis to date predicts that humans should have information to identify and reconstruct shape of an object in refractive stereo given that: i. Object is transparent and visible ii. Scene redirects incoming light just once and index of refraction is known. iii. Surface is both optically smooth and textured iv. At least two viewpoints should be available to obtain one 3D point on its light path at viewing surface. Psychophysical experiments that we are undertaking will either confirm or falsify human perceptual capability to identify and reconstruct shape of object under the above conditions.","annote":"31 May - 3 June 2016 at the University of Granada, Spain","author":[{"propositions":[],"lastnames":["Sultana"],"firstnames":["A."],"suffixes":[]},{"propositions":[],"lastnames":["Allison"],"firstnames":["R.","S."],"suffixes":[]}],"booktitle":"12th International Conference on Light and Color in Nature","date-added":"2016-12-04 21:56:23 +0000","date-modified":"2016-12-04 21:57:13 +0000","keywords":"Stereopsis","title":"Shape perception of water in photo-realistic 3D images","year":"2016","bibtex":"@incollection{Sultana:2016aa,\n\tabstract = {Light plays an extremely important role in the perception of transparency, depth and shape of water. In computer graphics, an important question is: How does perception of transparent objects (including 3D shape reconstruction) depend on the fidelity of the rendering. In this paper, I present a theoretical method to recover the surface shape of water under day light settings based on human visual stereoscopic view. Flat transparent objects with arbitrary depth should pass all the incoming light and we should obtain a clear view of the background through them. However, when the surface is not flat, the image of the background or the underlying surface gets distorted because the chromatic light passing through different regions of water surface experience distortion and absorption that varies with wavelength. The reflection and refraction angles of light hitting the surface of a material depend on the direction of the light, its spectral composition, the medium and its surface shape. In this study, we evaluate and improve the cues available for perceiving shape of refractive objects by exploring the relationship in a 3D view between (a) Reflective highlights on the water surface that depends largely on the lightning conditions and refractive features (with known index of refraction) seen through the surface of the medium, (b) viewing and perceiving conditions (stereoscopic and/or non stereoscopic) and (c) textures and shading that provide cues to distortions.\nAnalysis to date predicts that humans should have information to identify and reconstruct shape of an object in refractive stereo given that:\ni. Object is transparent and visible ii. Scene redirects incoming light just once and index of refraction is known.\niii. Surface is both optically smooth and textured iv. At least two viewpoints should be available to obtain one 3D point on its light path at viewing surface.\nPsychophysical experiments that we are undertaking will either confirm or falsify human perceptual capability to identify and reconstruct shape of object under the above conditions.},\n\tannote = {31 May - 3 June 2016 at the University of Granada, Spain},\n\tauthor = {Sultana, A. and Allison, R. S.},\n\tbooktitle = {12th International Conference on Light and Color in Nature},\n\tdate-added = {2016-12-04 21:56:23 +0000},\n\tdate-modified = {2016-12-04 21:57:13 +0000},\n\tkeywords = {Stereopsis},\n\ttitle = {Shape perception of water in photo-realistic 3D images},\n\tyear = {2016}}\n\n\n\n","author_short":["Sultana, A.","Allison, R. S."],"key":"Sultana:2016aa","id":"Sultana:2016aa","bibbaseid":"sultana-allison-shapeperceptionofwaterinphotorealistic3dimages-2016","role":"author","urls":{},"keyword":["Stereopsis"],"metadata":{"authorlinks":{"allison, r":"https://percept.eecs.yorku.ca/bibase%20pubs.shtml"}},"downloads":0},"search_terms":["shape","perception","water","photo","realistic","images","sultana","allison"],"keywords":["stereopsis"],"authorIDs":["vnY8GQ5AKXHNi7dqd"],"dataSources":["kmmXSosvtyJQxBtzs","BPKPSXjrbMGteC59J","MpMK4SvZzj5Fww5vJ","YbBWRH5Fc7xRr8ghk","szZaibkmSiiQBFQG8","DoyrDTpJ7HHCtki3q","JaoxzeTFRfvwgLoCW","XKwRm5Lx8Z9bzSzaP","AELuRZBpnp7nRDaqw"]}