Show simple item record

dc.contributor.advisorZickler, Todden_US
dc.contributor.authorXiong, Yingen_US
dc.date.accessioned2015-12-04T18:41:37Z
dc.date.created2015-11en_US
dc.date.issued2015-08-27en_US
dc.date.submitted2015en_US
dc.identifier.citationXiong, Ying. 2015. Physics-Based Visual Inference: Theory and Applications. Doctoral dissertation, Harvard University, Graduate School of Arts & Sciences.en_US
dc.identifier.urihttp://nrs.harvard.edu/urn-3:HUL.InstRepos:23845422
dc.description.abstractAnalyzing images to infer physical scene properties is a fundamental task in computer vision. It is by nature an ill-posed inverse problem, because imaging is a complicated, information-lossy physical and measurement process that cannot be deterministically inverted. This dissertation presents theory and algorithms for handling ambiguities in a variety of low-level vision problems. They are based on two key ideas: (1) explicitly modeling and reporting uncertainties are beneficial to visual inference; and (2) using local models can significantly reduce ambiguities that would exist in pixelwise analysis. In the first part of the dissertation, we study the color measurement pipeline of consumer digital cameras, and consider the inherent uncertainty of undoing the effects of tone-mapping. We introduce statistical models for this uncertainty and algorithms for fitting it to given cameras or imaging pipelines. Once fit, the model provides for each tone-mapped color a probability distribution over linear scene colors that could have induced it, which is demonstrated to be useful for a number of downstream inference applications. In the second part of the dissertation, we study the pixelwise ambiguities in physics-based visual inference and present theory and algorithms for employing local models to eliminate or reduce these ambiguities. In shape from shading, we perform mathematical analysis showing that when restricted with quadratic local models, the shape and lighting ambiguities will be reduced to a small finite number of choices as opposed to otherwise continuous manifolds. We propose a framework for surface reconstruction by enforcing consensus on the local regions, which is later enhanced and extended to be applicable to a variety of other visual inference tasks.en_US
dc.description.sponsorshipEngineering and Applied Sciences - Engineering Sciencesen_US
dc.format.mimetypeapplication/pdfen_US
dc.language.isoenen_US
dash.licenseLAAen_US
dc.subjectEngineering, Electronics and Electricalen_US
dc.subjectComputer Scienceen_US
dc.titlePhysics-Based Visual Inference: Theory and Applicationsen_US
dc.typeThesis or Dissertationen_US
dash.depositing.authorXiong, Yingen_US
dc.date.available2015-12-04T18:41:37Z
thesis.degree.date2015en_US
thesis.degree.grantorGraduate School of Arts & Sciencesen_US
thesis.degree.levelDoctoralen_US
thesis.degree.nameDoctor of Philosophyen_US
dc.contributor.committeeMemberGortler, Steven J.en_US
dc.contributor.committeeMemberLu, Yue M.en_US
dc.type.materialtexten_US
thesis.degree.departmentEngineering and Applied Sciences - Engineering Sciencesen_US
dash.identifier.vireohttp://etds.lib.harvard.edu/gsas/admin/view/587en_US
dc.description.keywordscomputer vision; visual inference; radiometric calibration; shape from shading; binocular stereoen_US
dash.author.emailying@xiongs.orgen_US
dash.identifier.drsurn-3:HUL.DRS.OBJECT:25142599en_US
dash.contributor.affiliatedXiong, Ying


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record