{"id":{"repo_id":"calpoly","oai_identifier":"oai:digitalcommons.calpoly.edu:theses-4912"},"canonical_url":"https://search.ndltd.org/etd/calpoly/oai:digitalcommons.calpoly.edu:theses-4912","repository":{"repo_id":"calpoly","name":"Cal Poly","base_url":"https://digitalcommons.calpoly.edu/do/oai/"},"display":{"title":"Video Generation Techniques for Novel View Synthesis with Flow-Matching Transformers","abstract":"Novel view synthesis (NVS) aims to generate images of a scene from unseen camera viewpoints. Recent work, such as Stable Virtual Camera, shows that large-scale image diffusion models like Stable Diffusion can be adapted for pose-conditioned view synthesis by incorporating video-generation techniques with camera conditioning. In this thesis, we introduce MVFlow, a new NVS model that extends this approach to a different image generation architecture: a flow-matching diffusion transformer, specifically FLUX.1, which has demonstrated strong performance in image synthesis. We evaluate MVFlow under varying input view counts and pose distance settings. Our results show that this architectural transfer is feasible; however, the current architecture and conditioning strategy do not yet match the performance of a mature baseline such as Stable Virtual Camera.","abstract_html":"Novel view synthesis (NVS) aims to generate images of a scene from unseen camera viewpoints. Recent work, such as Stable Virtual Camera, shows that large-scale image diffusion models like Stable Diffusion can be adapted for pose-conditioned view synthesis by incorporating video-generation techniques with camera conditioning. In this thesis, we introduce MVFlow, a new NVS model that extends this approach to a different image generation architecture: a flow-matching diffusion transformer, specifically FLUX.1, which has demonstrated strong performance in image synthesis. We evaluate MVFlow under varying input view counts and pose distance settings. Our results show that this architectural transfer is feasible; however, the current architecture and conditioning strategy do not yet match the performance of a mature baseline such as Stable Virtual Camera.","abstract_has_math":false,"creators":["Qiu, Xiuyuan"],"institution":null,"degree_name":"MS in Computer Science","degree_level":null,"degree_discipline":"Computer Science","degree_department":null,"school":null,"contributors":["Jonathan Ventura","Computer Science","College of Engineering"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-11-19T07:12:51Z","subjects":["Computer vision","novel view synthesis","video generation","diffusion","flow matching","Artificial Intelligence and Robotics"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[]},"links":{"outbound_url":"https://digitalcommons.calpoly.edu/theses/3246","outbound_label":"Repository record","outbound_source":"dc:identifier","record_url":"https://digitalcommons.calpoly.edu/theses/3246","record_label":"Repository record","record_kind":"record","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://digitalcommons.calpoly.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Adigitalcommons.calpoly.edu%3Atheses-4912","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Jonathan Ventura","Computer Science","College of Engineering"]},{"key":"dc:creator","label":"Author","values":["Qiu, Xiuyuan"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2026-03-20T07:00:00Z"]},{"key":"dc:date.created","label":"Dc Date Created","values":["2026-03-01T08:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Science"]},{"key":"thesis:degree_name","label":"Degree Name","values":["MS in Computer Science"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Computer vision","novel view synthesis","video generation","diffusion","flow matching","Artificial Intelligence and Robotics"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://digitalcommons.calpoly.edu/theses/3246"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Novel view synthesis (NVS) aims to generate images of a scene from unseen camera viewpoints. Recent work, such as Stable Virtual Camera, shows that large-scale image diffusion models like Stable Diffusion can be adapted for pose-conditioned view synthesis by incorporating video-generation techniques with camera conditioning. In this thesis, we introduce MVFlow, a new NVS model that extends this approach to a different image generation architecture: a flow-matching diffusion transformer, specifically FLUX.1, which has demonstrated strong performance in image synthesis. We evaluate MVFlow under varying input view counts and pose distance settings. Our results show that this architectural transfer is feasible; however, the current architecture and conditioning strategy do not yet match the performance of a mature baseline such as Stable Virtual Camera."]},{"key":"dc:title","label":"Title","values":["Video Generation Techniques for Novel View Synthesis with Flow-Matching Transformers"]}]}],"canonical_facts":{"dc:contributor":["Jonathan Ventura","Computer Science","College of Engineering"],"dc:creator":["Qiu, Xiuyuan"],"dc:date.available":["2026-03-20T07:00:00Z"],"dc:date.created":["2026-03-01T08:00:00Z"],"dc:description.abstract":["Novel view synthesis (NVS) aims to generate images of a scene from unseen camera viewpoints. Recent work, such as Stable Virtual Camera, shows that large-scale image diffusion models like Stable Diffusion can be adapted for pose-conditioned view synthesis by incorporating video-generation techniques with camera conditioning. In this thesis, we introduce MVFlow, a new NVS model that extends this approach to a different image generation architecture: a flow-matching diffusion transformer, specifically FLUX.1, which has demonstrated strong performance in image synthesis. We evaluate MVFlow under varying input view counts and pose distance settings. Our results show that this architectural transfer is feasible; however, the current architecture and conditioning strategy do not yet match the performance of a mature baseline such as Stable Virtual Camera."],"dc:identifier":["https://digitalcommons.calpoly.edu/theses/3246"],"dc:subject":["Computer vision","novel view synthesis","video generation","diffusion","flow matching","Artificial Intelligence and Robotics"],"dc:title":["Video Generation Techniques for Novel View Synthesis with Flow-Matching Transformers"],"thesis:degree_discipline":["Computer Science"],"thesis:degree_name":["MS in Computer Science"]},"updated_at":"2026-11-19T07:12:51Z"}