{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/127378"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/127378","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Deep feature manipulation for image synthesis and restoration","abstract":"Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2026-12-01","abstract_html":"Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2026-12-01","abstract_has_math":false,"creators":["Rojas Gomez, Renan Alfredo"],"institution":"University of Illinois at Urbana-Champaign","degree_name":"Ph.D.","degree_level":"Dissertation","degree_discipline":"Electrical & Computer Engr","degree_department":null,"school":null,"contributors":["Do, Minh N","Boppart, Stephen A","Gupta, Saurabh","Nguyen, Anh M","Zhao, Zhizhen"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2024,"date_issued":"2024-12-02","date_published":"2024-12-02","updated_at":"2026-07-22T22:25:04Z","subjects":["Image Synthesis","Image Restoration","Computer Vision","Deep Learning","Signal Processing","Convolutional Neural Networks","Vision Transformers","Multiscale Image Representation"],"languages":["en","eng"],"rights":["Copyright 2024 Renan Rojas Gomez"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/2142/127378","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Do, Minh N","Boppart, Stephen A","Gupta, Saurabh","Nguyen, Anh M","Zhao, Zhizhen"]},{"key":"dc:creator","label":"Author","values":["Rojas Gomez, Renan Alfredo"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2024-12-02","2024-12"]},{"key":"dc:type","label":"Dc Type","values":["text","Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Electrical & Computer Engr"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Dissertation"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Ph.D."]},{"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":["Image Synthesis","Image Restoration","Computer Vision","Deep Learning","Signal Processing","Convolutional Neural Networks","Vision Transformers","Multiscale Image Representation"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en","eng"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2024 Renan Rojas Gomez"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://hdl.handle.net/2142/127378"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2026-12-01","The student, Renan Rojas Gomez, accepted the attached license on 2024-11-30 at 14:43.","The student, Renan Rojas Gomez, submitted this Dissertation for approval on 2024-11-30 at 15:05.","This Dissertation was approved for publication on 2024-12-02 at 09:50.","DSpace SAF Submission Ingestion Package generated from Vireo submission #21428 on 2025-03-28 at 14:43:33","Deep learning has revolutionized image generation by capturing the underlying statistical distribution of training data, allowing for the creation of new, realistic images through latent representation manipulation. Deep learning has also significantly advanced image restoration techniques by improving the recovery of degraded images via removing noise, improving contrast, and enhancing details more effectively than traditional techniques, leading to state-of-the-art performance. Nevertheless, deep learning's success in image synthesis and restoration is often hampered by its neglect of data properties, model architecture, and task specifics. For instance, ignoring spatial invariances in tasks like image classification and segmentation can lead to models vulnerable to minor spatial transformations. Latent representations that are not interpretable can limit the model's ability to control salient image attributes, while the lack of feature invertibility can hinder high-quality image generation. To address these issues, this work introduces novel techniques that enhance performance by considering spatial invariances, capturing meaningful features, and improving feature invertibility. We propose (i) a novel down and upsampling method in CNN models to impose perfect shift equivariance towards improved performance on image classification and segmentation, (ii) a data-adaptive framework to build provably shift-equivariant ViT models, improving performance on discriminative tasks across hierarchical architectures, (iii) a neural style transfer-based data-augmentation technique to extract strong image representations through self-supervised learning, (iv) a computationally efficient photorealistic style transfer algorithm that matches multiscale geometric representations, eliminating the need for learned features to produce high-quality results by preserving fine-grained visual details, and (v) the use of adversarially robust features as a perceptual prior to invert contracted representations back into high-quality images, outperforming standard models and unlocking their use in various image synthesis and restoration tasks."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Deep feature manipulation for image synthesis and restoration"]}]}],"canonical_facts":{"dc:contributor":["Do, Minh N","Boppart, Stephen A","Gupta, Saurabh","Nguyen, Anh M","Zhao, Zhizhen"],"dc:creator":["Rojas Gomez, Renan Alfredo"],"dc:date":["2024-12-02","2024-12"],"dc:description":["Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2026-12-01","The student, Renan Rojas Gomez, accepted the attached license on 2024-11-30 at 14:43.","The student, Renan Rojas Gomez, submitted this Dissertation for approval on 2024-11-30 at 15:05.","This Dissertation was approved for publication on 2024-12-02 at 09:50.","DSpace SAF Submission Ingestion Package generated from Vireo submission #21428 on 2025-03-28 at 14:43:33","Deep learning has revolutionized image generation by capturing the underlying statistical distribution of training data, allowing for the creation of new, realistic images through latent representation manipulation. Deep learning has also significantly advanced image restoration techniques by improving the recovery of degraded images via removing noise, improving contrast, and enhancing details more effectively than traditional techniques, leading to state-of-the-art performance. Nevertheless, deep learning's success in image synthesis and restoration is often hampered by its neglect of data properties, model architecture, and task specifics. For instance, ignoring spatial invariances in tasks like image classification and segmentation can lead to models vulnerable to minor spatial transformations. Latent representations that are not interpretable can limit the model's ability to control salient image attributes, while the lack of feature invertibility can hinder high-quality image generation. To address these issues, this work introduces novel techniques that enhance performance by considering spatial invariances, capturing meaningful features, and improving feature invertibility. We propose (i) a novel down and upsampling method in CNN models to impose perfect shift equivariance towards improved performance on image classification and segmentation, (ii) a data-adaptive framework to build provably shift-equivariant ViT models, improving performance on discriminative tasks across hierarchical architectures, (iii) a neural style transfer-based data-augmentation technique to extract strong image representations through self-supervised learning, (iv) a computationally efficient photorealistic style transfer algorithm that matches multiscale geometric representations, eliminating the need for learned features to produce high-quality results by preserving fine-grained visual details, and (v) the use of adversarially robust features as a perceptual prior to invert contracted representations back into high-quality images, outperforming standard models and unlocking their use in various image synthesis and restoration tasks."],"dc:format":["application/pdf"],"dc:identifier":["https://hdl.handle.net/2142/127378"],"dc:language":["en","eng"],"dc:rights":["Copyright 2024 Renan Rojas Gomez"],"dc:subject":["Image Synthesis","Image Restoration","Computer Vision","Deep Learning","Signal Processing","Convolutional Neural Networks","Vision Transformers","Multiscale Image Representation"],"dc:title":["Deep feature manipulation for image synthesis and restoration"],"dc:type":["text","Thesis"],"thesis:degree_discipline":["Electrical & Computer Engr"],"thesis:degree_level":["Dissertation"],"thesis:degree_name":["Ph.D."],"thesis:institution_name":["University of Illinois at Urbana-Champaign"]},"updated_at":"2026-07-22T22:25:04Z"}