Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/181882
Full metadata record
DC FieldValueLanguage
dc.contributor.authorFu, Honghaoen_US
dc.date.accessioned2024-12-27T13:31:16Z-
dc.date.available2024-12-27T13:31:16Z-
dc.date.issued2024-
dc.identifier.citationFu, H. (2024). In-the-wild image quality assessment with diffusion priors. Master's thesis, Nanyang Technological University, Singapore. https://hdl.handle.net/10356/181882en_US
dc.identifier.urihttps://hdl.handle.net/10356/181882-
dc.description.abstractBlind image quality assessment (IQA) in the wild, which assesses the quality of images with complex authentic distortions and no reference images, presents significant challenges. Given the difficulty in collecting large-scale training data, leveraging limited data to develop a model with strong generalization remains an open problem. Motivated by the robust image perception capabilities of pretrained text-to image (T2I) diffusion models, we propose a novel IQA method, diffusion priors-based IQA (DP-IQA), to utilize the T2I model’s prior for improved performance and generalization ability. Specifically, we utilize pre-trained Stable Diffusion as the backbone, extracting multi-level features from the denoising U-Net guided by prompt embeddings through a tunable text adapter. Simultaneously, an image adapter compensates for information loss introduced by the lossy pre-trained encoder. Unlike T2I models that require full image distribution modeling, our approach targets image quality assessment, which inherently requires fewer parameters. To improve applicability, we distill the knowledge into a lightweight CNN-based student model, significantly reducing parameters while maintaining or even enhancing generalization performance. Experimental results demonstrate that DP-IQA achieves state-of-the-art performance on various in-the-wild datasets, highlighting the superior generalization capability of T2I priors in blind IQA tasks. To our knowledge, DP-IQA is the first method to apply pre-trained diffusion priors in blind IQA.en_US
dc.language.isoenen_US
dc.publisherNanyang Technological Universityen_US
dc.subjectComputer and Information Scienceen_US
dc.titleIn-the-wild image quality assessment with diffusion priorsen_US
dc.typeThesis-Master by Courseworken_US
dc.contributor.supervisorWen Bihanen_US
dc.contributor.schoolSchool of Electrical and Electronic Engineeringen_US
dc.description.degreeMaster's degreeen_US
dc.contributor.researchRapid-Rich Object Search (ROSE) Laben_US
dc.contributor.supervisoremailbihan.wen@ntu.edu.sgen_US
dc.subject.keywordsBlind IQAen_US
dc.subject.keywordsDiffusion prioren_US
dc.subject.keywordsText-to-image modelen_US
dc.subject.keywordsKnowledge distillationen_US
item.fulltextWith Fulltext-
item.grantfulltextrestricted-
Appears in Collections:EEE Theses
Files in This Item:
File Description SizeFormat 
Fu Honghao-Dissertation_Final_version.pdf
  Restricted Access
2.8 MBAdobe PDFView/Open

Page view(s)

77
Updated on Feb 10, 2025

Download(s)

1
Updated on Feb 10, 2025

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.