{"id":{"repo_id":"mcmaster","oai_identifier":"oai:macsphere.mcmaster.ca:11375/32258"},"canonical_url":"https://search.ndltd.org/etd/mcmaster/oai:macsphere.mcmaster.ca:11375/32258","repository":{"repo_id":"mcmaster","name":"McMaster University","base_url":"https://macsphere.mcmaster.ca/server/oai/request"},"display":{"title":"Prior-guided Neural Compression of Visual Data","abstract":"In recent years, deep learning methods have been widely applied in the ﬁeld of visual data compression, some of which have delivered the best rate-distortion performances in history. The target domain of neural data compression started from that of still images and then rapidly expanded into various modalities, including videos, point clouds, and the emerging 3D Gaussian Splatting (3DGS) representations. However, steady progresses of the neural compression research aside, some technical challenges still remain. One of them is the high computational costs of current neural compression models. Although more complex neural network architectures often yield improved compression performance, the marginal beneﬁt of blindly increasing model complexity is diminishing. For most applications the rapidly increased cost cannot be justiﬁed for ever small coding gains. Unless better cost-performance ratio is achieved, neural visual data compression methods are unlikely to replace traditional compression standards that are deeply entrenched in real world. To address this issue, we advocate to incorporate known priors, such as signal sparsity and eﬃcient space covering structures in quantization, into the design of neural compression architecture. Compared with the mainstream pure big data-driven black box learning approach, our prior-guided design approach can lead to signiﬁcant coding gains with no or very small overheads in either model size or computational complexity, improving cost-eﬀectiveness in practical deployment. With such motivations we design, implement and experiment with a series of neural compression models for three visual data modalities: images, 3D point clouds, and 3DGS scene representations. Extensive experimental results demonstrate that our prior-guided neural compression models can deliver rate-distortion performance comparable to state-of-the-art methods at a much reduced resource level. In future research, we plan to further optimize the proposed paradigm to discover even more eﬃcient and practical neural compression models, and in parallel we will expand its applications to other data modalities, such as volumetric video and panoramic video.","abstract_html":"In recent years, deep learning methods have been widely applied in the ﬁeld of visual data compression, some of which have delivered the best rate-distortion performances in history. The target domain of neural data compression started from that of still images and then rapidly expanded into various modalities, including videos, point clouds, and the emerging 3D Gaussian Splatting (3DGS) representations. However, steady progresses of the neural compression research aside, some technical challenges still remain. One of them is the high computational costs of current neural compression models. Although more complex neural network architectures often yield improved compression performance, the marginal beneﬁt of blindly increasing model complexity is diminishing. For most applications the rapidly increased cost cannot be justiﬁed for ever small coding gains. Unless better cost-performance ratio is achieved, neural visual data compression methods are unlikely to replace traditional compression standards that are deeply entrenched in real world. To address this issue, we advocate to incorporate known priors, such as signal sparsity and eﬃcient space covering structures in quantization, into the design of neural compression architecture. Compared with the mainstream pure big data-driven black box learning approach, our prior-guided design approach can lead to signiﬁcant coding gains with no or very small overheads in either model size or computational complexity, improving cost-eﬀectiveness in practical deployment. With such motivations we design, implement and experiment with a series of neural compression models for three visual data modalities: images, 3D point clouds, and 3DGS scene representations. Extensive experimental results demonstrate that our prior-guided neural compression models can deliver rate-distortion performance comparable to state-of-the-art methods at a much reduced resource level. In future research, we plan to further optimize the proposed paradigm to discover even more eﬃcient and practical neural compression models, and in parallel we will expand its applications to other data modalities, such as volumetric video and panoramic video.","abstract_has_math":false,"creators":["Xu, Hao"],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":"Electrical and Computer Engineering","school":null,"contributors":[],"advisors":["Wu, Xiaolin"],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025","date_published":"2025","updated_at":"2026-11-18T19:32:06Z","subjects":[],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://doi.org/10.71548/11126"],"render_values":[{"text":"https://doi.org/10.71548/11126","href":"https://doi.org/10.71548/11126","code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/11375/32258","outbound_label":"Handle","outbound_source":"dc:identifier.uri","record_url":"http://hdl.handle.net/11375/32258","record_label":"Repository record","record_kind":"handle","record_source":"dc:identifier.uri","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://macsphere.mcmaster.ca/server/oai/request?verb=GetRecord&metadataPrefix=dim&identifier=oai%3Amacsphere.mcmaster.ca%3A11375%2F32258","prefix":"dim"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Wu, Xiaolin"]},{"key":"dc:contributor.department","label":"Department","values":["Electrical and Computer Engineering"]},{"key":"dc:creator","label":"Author","values":["Xu, Hao"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2025-08-28T15:17:22Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2025-08-28T15:17:22Z"]},{"key":"dc:date.issued","label":"Date","values":["2025"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["http://hdl.handle.net/11375/32258","https://doi.org/10.71548/11126"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["In recent years, deep learning methods have been widely applied in the ﬁeld of visual data compression, some of which have delivered the best rate-distortion performances in history. The target domain of neural data compression started from that of still images and then rapidly expanded into various modalities, including videos, point clouds, and the emerging 3D Gaussian Splatting (3DGS) representations. However, steady progresses of the neural compression research aside, some technical challenges still remain. One of them is the high computational costs of current neural compression models. Although more complex neural network architectures often yield improved compression performance, the marginal beneﬁt of blindly increasing model complexity is diminishing. For most applications the rapidly increased cost cannot be justiﬁed for ever small coding gains. Unless better cost-performance ratio is achieved, neural visual data compression methods are unlikely to replace traditional compression standards that are deeply entrenched in real world. To address this issue, we advocate to incorporate known priors, such as signal sparsity and eﬃcient space covering structures in quantization, into the design of neural compression architecture. Compared with the mainstream pure big data-driven black box learning approach, our prior-guided design approach can lead to signiﬁcant coding gains with no or very small overheads in either model size or computational complexity, improving cost-eﬀectiveness in practical deployment. With such motivations we design, implement and experiment with a series of neural compression models for three visual data modalities: images, 3D point clouds, and 3DGS scene representations. Extensive experimental results demonstrate that our prior-guided neural compression models can deliver rate-distortion performance comparable to state-of-the-art methods at a much reduced resource level. In future research, we plan to further optimize the proposed paradigm to discover even more eﬃcient and practical neural compression models, and in parallel we will expand its applications to other data modalities, such as volumetric video and panoramic video."]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["Doctor of Philosophy (PhD)"]},{"key":"dc:title","label":"Title","values":["Prior-guided Neural Compression of Visual Data"]},{"key":"others:access_status","label":"Others Access Status","values":["open.access"]}]}],"canonical_facts":{"dc:contributor.advisor":["Wu, Xiaolin"],"dc:contributor.department":["Electrical and Computer Engineering"],"dc:creator":["Xu, Hao"],"dc:date.accessioned":["2025-08-28T15:17:22Z"],"dc:date.available":["2025-08-28T15:17:22Z"],"dc:date.issued":["2025"],"dc:description.abstract":["In recent years, deep learning methods have been widely applied in the ﬁeld of visual data compression, some of which have delivered the best rate-distortion performances in history. The target domain of neural data compression started from that of still images and then rapidly expanded into various modalities, including videos, point clouds, and the emerging 3D Gaussian Splatting (3DGS) representations. However, steady progresses of the neural compression research aside, some technical challenges still remain. One of them is the high computational costs of current neural compression models. Although more complex neural network architectures often yield improved compression performance, the marginal beneﬁt of blindly increasing model complexity is diminishing. For most applications the rapidly increased cost cannot be justiﬁed for ever small coding gains. Unless better cost-performance ratio is achieved, neural visual data compression methods are unlikely to replace traditional compression standards that are deeply entrenched in real world. To address this issue, we advocate to incorporate known priors, such as signal sparsity and eﬃcient space covering structures in quantization, into the design of neural compression architecture. Compared with the mainstream pure big data-driven black box learning approach, our prior-guided design approach can lead to signiﬁcant coding gains with no or very small overheads in either model size or computational complexity, improving cost-eﬀectiveness in practical deployment. With such motivations we design, implement and experiment with a series of neural compression models for three visual data modalities: images, 3D point clouds, and 3DGS scene representations. Extensive experimental results demonstrate that our prior-guided neural compression models can deliver rate-distortion performance comparable to state-of-the-art methods at a much reduced resource level. In future research, we plan to further optimize the proposed paradigm to discover even more eﬃcient and practical neural compression models, and in parallel we will expand its applications to other data modalities, such as volumetric video and panoramic video."],"dc:description.degree":["Doctor of Philosophy (PhD)"],"dc:identifier.uri":["http://hdl.handle.net/11375/32258","https://doi.org/10.71548/11126"],"dc:language.iso":["en"],"dc:title":["Prior-guided Neural Compression of Visual Data"],"dc:type":["Thesis"],"others:access_status":["open.access"]},"updated_at":"2026-11-18T19:32:06Z"}