Untrained Neural Network Priors for Inverse Imaging Problems: A Survey
<p dir="ltr">In recent years, advancements in machine learning (ML) techniques, in particular, deep learning (DL) methods have gained a lot of momentum in solving inverse imaging problems, often surpassing the performance provided by hand-crafted approaches. Traditionally, analytical...
محفوظ في:
| المؤلف الرئيسي: | |
|---|---|
| مؤلفون آخرون: | , , , , |
| منشور في: |
2022
|
| الموضوعات: | |
| الوسوم: |
إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
|
| _version_ | 1864513560974983168 |
|---|---|
| author | Adnan Qayyum (16875936) |
| author2 | Inaam Ilahi (16904676) Fahad Shamshad (16904679) Farid Boussaid (5334431) Mohammed Bennamoun (6883301) Junaid Qadir (16494902) |
| author2_role | author author author author author |
| author_facet | Adnan Qayyum (16875936) Inaam Ilahi (16904676) Fahad Shamshad (16904679) Farid Boussaid (5334431) Mohammed Bennamoun (6883301) Junaid Qadir (16494902) |
| author_role | author |
| dc.creator.none.fl_str_mv | Adnan Qayyum (16875936) Inaam Ilahi (16904676) Fahad Shamshad (16904679) Farid Boussaid (5334431) Mohammed Bennamoun (6883301) Junaid Qadir (16494902) |
| dc.date.none.fl_str_mv | 2022-09-05T00:00:00Z |
| dc.identifier.none.fl_str_mv | 10.1109/tpami.2022.3204527 |
| dc.relation.none.fl_str_mv | https://figshare.com/articles/journal_contribution/Untrained_Neural_Network_Priors_for_Inverse_Imaging_Problems_A_Survey/24056355 |
| dc.rights.none.fl_str_mv | CC BY 4.0 info:eu-repo/semantics/openAccess |
| dc.subject.none.fl_str_mv | Information and computing sciences Machine learning Inverse problems Neural networks Imaging Task analysis Image reconstruction Noise measurement Deep learning Inverse imaging problems Untrained neural networks priors |
| dc.title.none.fl_str_mv | Untrained Neural Network Priors for Inverse Imaging Problems: A Survey |
| dc.type.none.fl_str_mv | Text Journal contribution info:eu-repo/semantics/publishedVersion text contribution to journal |
| description | <p dir="ltr">In recent years, advancements in machine learning (ML) techniques, in particular, deep learning (DL) methods have gained a lot of momentum in solving inverse imaging problems, often surpassing the performance provided by hand-crafted approaches. Traditionally, analytical methods have been used to solve inverse imaging problems such as image restoration, inpainting, and superresolution. Unlike analytical methods for which the problem is explicitly defined and the domain knowledge is carefully engineered into the solution, DL models do not benefit from such prior knowledge and instead make use of large datasets to predict an unknown solution to the inverse problem. Recently, a new paradigm of training deep models using a single image, named untrained neural network prior (UNNP) has been proposed to solve a variety of inverse tasks, e.g., restoration and inpainting. Since then, many researchers have proposed various applications and variants of UNNP. In this paper, we present a comprehensive review of such studies and various UNNP applications for different tasks and highlight various open research problems which require further research.</p><h2>Other Information</h2><p dir="ltr">Published in: IEEE Transactions on Pattern Analysis and Machine Intelligence<br>License: <a href="https://creativecommons.org/licenses/by/4.0/legalcode" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1109/tpami.2022.3204527" target="_blank">https://dx.doi.org/10.1109/tpami.2022.3204527</a></p> |
| eu_rights_str_mv | openAccess |
| id | Manara2_7350f17f8dab7ab75eeb5fb9f987c436 |
| identifier_str_mv | 10.1109/tpami.2022.3204527 |
| network_acronym_str | Manara2 |
| network_name_str | Manara2 |
| oai_identifier_str | oai:figshare.com:article/24056355 |
| publishDate | 2022 |
| repository.mail.fl_str_mv | |
| repository.name.fl_str_mv | |
| repository_id_str | |
| rights_invalid_str_mv | CC BY 4.0 |
| spelling | Untrained Neural Network Priors for Inverse Imaging Problems: A SurveyAdnan Qayyum (16875936)Inaam Ilahi (16904676)Fahad Shamshad (16904679)Farid Boussaid (5334431)Mohammed Bennamoun (6883301)Junaid Qadir (16494902)Information and computing sciencesMachine learningInverse problemsNeural networksImagingTask analysisImage reconstructionNoise measurementDeep learningInverse imaging problemsUntrained neural networks priors<p dir="ltr">In recent years, advancements in machine learning (ML) techniques, in particular, deep learning (DL) methods have gained a lot of momentum in solving inverse imaging problems, often surpassing the performance provided by hand-crafted approaches. Traditionally, analytical methods have been used to solve inverse imaging problems such as image restoration, inpainting, and superresolution. Unlike analytical methods for which the problem is explicitly defined and the domain knowledge is carefully engineered into the solution, DL models do not benefit from such prior knowledge and instead make use of large datasets to predict an unknown solution to the inverse problem. Recently, a new paradigm of training deep models using a single image, named untrained neural network prior (UNNP) has been proposed to solve a variety of inverse tasks, e.g., restoration and inpainting. Since then, many researchers have proposed various applications and variants of UNNP. In this paper, we present a comprehensive review of such studies and various UNNP applications for different tasks and highlight various open research problems which require further research.</p><h2>Other Information</h2><p dir="ltr">Published in: IEEE Transactions on Pattern Analysis and Machine Intelligence<br>License: <a href="https://creativecommons.org/licenses/by/4.0/legalcode" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1109/tpami.2022.3204527" target="_blank">https://dx.doi.org/10.1109/tpami.2022.3204527</a></p>2022-09-05T00:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1109/tpami.2022.3204527https://figshare.com/articles/journal_contribution/Untrained_Neural_Network_Priors_for_Inverse_Imaging_Problems_A_Survey/24056355CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/240563552022-09-05T00:00:00Z |
| spellingShingle | Untrained Neural Network Priors for Inverse Imaging Problems: A Survey Adnan Qayyum (16875936) Information and computing sciences Machine learning Inverse problems Neural networks Imaging Task analysis Image reconstruction Noise measurement Deep learning Inverse imaging problems Untrained neural networks priors |
| status_str | publishedVersion |
| title | Untrained Neural Network Priors for Inverse Imaging Problems: A Survey |
| title_full | Untrained Neural Network Priors for Inverse Imaging Problems: A Survey |
| title_fullStr | Untrained Neural Network Priors for Inverse Imaging Problems: A Survey |
| title_full_unstemmed | Untrained Neural Network Priors for Inverse Imaging Problems: A Survey |
| title_short | Untrained Neural Network Priors for Inverse Imaging Problems: A Survey |
| title_sort | Untrained Neural Network Priors for Inverse Imaging Problems: A Survey |
| topic | Information and computing sciences Machine learning Inverse problems Neural networks Imaging Task analysis Image reconstruction Noise measurement Deep learning Inverse imaging problems Untrained neural networks priors |