{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/49707"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/49707","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Model-based metrics for assessing completeness and accuracy for 3D image-based reconstruction methods","abstract":"Structure from Motion (SfM) is a technique for inferring the underlying 3D geometry of the real world from a set of unordered images using a pinhole camera model. Due to a wider availability of cameras, the availability of commodity smart phones and tablets equipped with cameras, and enhanced computational power, more attention is now being given to leverage visual data such as still images and video streams for the automated processes such as mobile augmented reality and engineering performance monitoring applications. Discuss significant research progress related to leveraging rich visual data for 3D image-based point butt modeling purposes, today there is no way to provide feedback to users who are taken these imagery on the best location and viewpoints for cameras. To address current limitations, this thesis introduces 3D image-based modeling incrementation -- a variant of incremental SfM -- where no prior knowledge of the new image/cameras is given. The state-of-the-art online algorithm for SfM only focus on processing video streams in which the images are ordered and sequence is known as a priori. In this thesis, we use a heuristic approach to avoid the need for pair-wise image matching which is a bottleneck for incremental SfM processes. We also discuss how we measure the completeness of the 3D point butt without the knowledge of the ground truth. We propose a general framework for feedback systems where the location and the pose of the new camera that can enhance the completeness of the 3D image-based point butt is predicted. We demonstrate that this framework can assist users of SfM methods to automatically improve the completeness of the reconstructed 3D point butt models form the real world scenery. We validate our algorithm on both small and large dataset as well as indoor and outdoor scenes using daily photologs collected from two ongoing construction sites. The perceived benefits and practical significance of the proposed method are also discussed in detail.","abstract_html":"Structure from Motion (SfM) is a technique for inferring the underlying 3D geometry of the real world from a set of unordered images using a pinhole camera model. Due to a wider availability of cameras, the availability of commodity smart phones and tablets equipped with cameras, and enhanced computational power, more attention is now being given to leverage visual data such as still images and video streams for the automated processes such as mobile augmented reality and engineering performance monitoring applications. Discuss significant research progress related to leveraging rich visual data for 3D image-based point butt modeling purposes, today there is no way to provide feedback to users who are taken these imagery on the best location and viewpoints for cameras. To address current limitations, this thesis introduces 3D image-based modeling incrementation -- a variant of incremental SfM -- where no prior knowledge of the new image/cameras is given. The state-of-the-art online algorithm for SfM only focus on processing video streams in which the images are ordered and sequence is known as a priori. In this thesis, we use a heuristic approach to avoid the need for pair-wise image matching which is a bottleneck for incremental SfM processes. We also discuss how we measure the completeness of the 3D point butt without the knowledge of the ground truth. We propose a general framework for feedback systems where the location and the pose of the new camera that can enhance the completeness of the 3D image-based point butt is predicted. We demonstrate that this framework can assist users of SfM methods to automatically improve the completeness of the reconstructed 3D point butt models form the real world scenery. We validate our algorithm on both small and large dataset as well as indoor and outdoor scenes using daily photologs collected from two ongoing construction sites. The perceived benefits and practical significance of the proposed method are also discussed in detail.","abstract_has_math":false,"creators":["Rukkanchanunt, Thapanapong"],"institution":"University of Illinois at Urbana-Champaign","degree_name":"M.S.","degree_level":"Thesis","degree_discipline":"Computer Science","degree_department":null,"school":null,"contributors":["Golparvar-Fard, Mani"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2014,"date_issued":"2014-05-30T17:05:48Z","date_published":"2014-05-30T17:05:48Z","updated_at":"2026-07-22T22:25:38Z","subjects":["Structure from Motion","Point butt","Completeness"],"languages":["en"],"rights":["Copyright 2014 Thapanapong Rukkanchanunt"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/2142/49707","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Golparvar-Fard, Mani"]},{"key":"dc:creator","label":"Author","values":["Rukkanchanunt, Thapanapong"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2014-05-30T17:05:48Z","2016-09-22T20:59:12Z","2014-05"]},{"key":"dc:type","label":"Dc Type","values":["text"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Science"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["M.S."]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of Illinois at Urbana-Champaign"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Structure from Motion","Point butt","Completeness"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2014 Thapanapong Rukkanchanunt"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["http://hdl.handle.net/2142/49707"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Structure from Motion (SfM) is a technique for inferring the underlying 3D geometry of the real world from a set of unordered images using a pinhole camera model. Due to a wider availability of cameras, the availability of commodity smart phones and tablets equipped with cameras, and enhanced computational power, more attention is now being given to leverage visual data such as still images and video streams for the automated processes such as mobile augmented reality and engineering performance monitoring applications. Discuss significant research progress related to leveraging rich visual data for 3D image-based point butt modeling purposes, today there is no way to provide feedback to users who are taken these imagery on the best location and viewpoints for cameras. To address current limitations, this thesis introduces 3D image-based modeling incrementation -- a variant of incremental SfM -- where no prior knowledge of the new image/cameras is given. The state-of-the-art online algorithm for SfM only focus on processing video streams in which the images are ordered and sequence is known as a priori. In this thesis, we use a heuristic approach to avoid the need for pair-wise image matching which is a bottleneck for incremental SfM processes. We also discuss how we measure the completeness of the 3D point butt without the knowledge of the ground truth. We propose a general framework for feedback systems where the location and the pose of the new camera that can enhance the completeness of the 3D image-based point butt is predicted. We demonstrate that this framework can assist users of SfM methods to automatically improve the completeness of the reconstructed 3D point butt models form the real world scenery. We validate our algorithm on both small and large dataset as well as indoor and outdoor scenes using daily photologs collected from two ongoing construction sites. The perceived benefits and practical significance of the proposed method are also discussed in detail.","Item withdrawn by Laura Spradlin (lspradl2@illinois.edu) on 2014-05-01T18:58:12Z Item was in collections: University of Illinois Theses & Dissertations (ID: 1) No. of bitstreams: 1 Rukkanchanunt_Thapanapong.pdf: 10117196 bytes, checksum: 296b4cbb2399ddf1ba0f66bfee8d7a7c (MD5)","Made available in DSpace on 2014-05-30T17:05:48Z (GMT). No. of bitstreams: 2 Thapanapong_Rukkanchanunt.pdf: 10117462 bytes, checksum: 5134cbcc7b924b2b2f8f67b8b4739ae7 (MD5) license.txt: 4075 bytes, checksum: 80efbe7957a75189417188b6598c9043 (MD5)","Item marked as restricted to the 'UIUC Users [automated]' Group (id=2) by Seth Robbins (robbins.sd@gmail.com) on 2014-05-30T17:09:45Z Item is restricted until 2016-05-30T17:09:03Z","Restriction data tranferred 2014-07-01T11:38:50-05:00 Original Data Group with Access UIUC Users [automated] Release Date: 2016-05-30 12:09:03 UTC Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD system","U of I Only Restriction Lifted for Item 49758 on 2016-09-22T20:59:12Z."]},{"key":"dc:title","label":"Title","values":["Model-based metrics for assessing completeness and accuracy for 3D image-based reconstruction methods"]}]}],"canonical_facts":{"dc:contributor":["Golparvar-Fard, Mani"],"dc:creator":["Rukkanchanunt, Thapanapong"],"dc:date":["2014-05-30T17:05:48Z","2016-09-22T20:59:12Z","2014-05"],"dc:description":["Structure from Motion (SfM) is a technique for inferring the underlying 3D geometry of the real world from a set of unordered images using a pinhole camera model. Due to a wider availability of cameras, the availability of commodity smart phones and tablets equipped with cameras, and enhanced computational power, more attention is now being given to leverage visual data such as still images and video streams for the automated processes such as mobile augmented reality and engineering performance monitoring applications. Discuss significant research progress related to leveraging rich visual data for 3D image-based point butt modeling purposes, today there is no way to provide feedback to users who are taken these imagery on the best location and viewpoints for cameras. To address current limitations, this thesis introduces 3D image-based modeling incrementation -- a variant of incremental SfM -- where no prior knowledge of the new image/cameras is given. The state-of-the-art online algorithm for SfM only focus on processing video streams in which the images are ordered and sequence is known as a priori. In this thesis, we use a heuristic approach to avoid the need for pair-wise image matching which is a bottleneck for incremental SfM processes. We also discuss how we measure the completeness of the 3D point butt without the knowledge of the ground truth. We propose a general framework for feedback systems where the location and the pose of the new camera that can enhance the completeness of the 3D image-based point butt is predicted. We demonstrate that this framework can assist users of SfM methods to automatically improve the completeness of the reconstructed 3D point butt models form the real world scenery. We validate our algorithm on both small and large dataset as well as indoor and outdoor scenes using daily photologs collected from two ongoing construction sites. The perceived benefits and practical significance of the proposed method are also discussed in detail.","Item withdrawn by Laura Spradlin (lspradl2@illinois.edu) on 2014-05-01T18:58:12Z Item was in collections: University of Illinois Theses & Dissertations (ID: 1) No. of bitstreams: 1 Rukkanchanunt_Thapanapong.pdf: 10117196 bytes, checksum: 296b4cbb2399ddf1ba0f66bfee8d7a7c (MD5)","Made available in DSpace on 2014-05-30T17:05:48Z (GMT). No. of bitstreams: 2 Thapanapong_Rukkanchanunt.pdf: 10117462 bytes, checksum: 5134cbcc7b924b2b2f8f67b8b4739ae7 (MD5) license.txt: 4075 bytes, checksum: 80efbe7957a75189417188b6598c9043 (MD5)","Item marked as restricted to the 'UIUC Users [automated]' Group (id=2) by Seth Robbins (robbins.sd@gmail.com) on 2014-05-30T17:09:45Z Item is restricted until 2016-05-30T17:09:03Z","Restriction data tranferred 2014-07-01T11:38:50-05:00 Original Data Group with Access UIUC Users [automated] Release Date: 2016-05-30 12:09:03 UTC Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD system","U of I Only Restriction Lifted for Item 49758 on 2016-09-22T20:59:12Z."],"dc:identifier":["http://hdl.handle.net/2142/49707"],"dc:language":["en"],"dc:rights":["Copyright 2014 Thapanapong Rukkanchanunt"],"dc:subject":["Structure from Motion","Point butt","Completeness"],"dc:title":["Model-based metrics for assessing completeness and accuracy for 3D image-based reconstruction methods"],"dc:type":["text"],"thesis:degree_discipline":["Computer Science"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["M.S."],"thesis:institution_name":["University of Illinois at Urbana-Champaign"]},"updated_at":"2026-07-22T22:25:38Z"}