Filters








58 Hits in 3.7 sec

RankIQA: Learning from Rankings for No-reference Image Quality Assessment [article]

Xialei Liu, Joost van de Weijer, Andrew D. Bagdanov
2017 arXiv   pre-print
We propose a no-reference image quality assessment (NR-IQA) approach that learns from rankings (RankIQA).  ...  To address the problem of limited IQA dataset size, we train a Siamese Network to rank images in terms of image quality by using synthetically generated distortions for which relative image quality is  ...  We call this learning from rankings approach RankIQA, and with it we learn to rank image in terms of quality using Siamese Networks, and then we transfer knowledge learned from ranked images to a traditional  ... 
arXiv:1707.08347v1 fatcat:vm62nereqvhqzl3wfxydmkz6ci

Exploiting Unlabeled Data in CNNs by Self-supervised Learning to Rank

Xialei Liu, Joost Van De Weijer, Andrew D. Bagdanov
2019 IEEE Transactions on Pattern Analysis and Machine Intelligence  
We apply our framework to two regression problems: Image Quality Assessment (IQA) and Crowd Counting. For both we show how to automatically generate ranked image sets from unlabeled data.  ...  Our results show that networks trained to regress to the ground truth targets for labeled data and to simultaneously learn to rank unlabeled data obtain significantly better, state-of-the-art results for  ...  Xialei Liu acknowledges the Chinese Scholarship Council (CSC) grant No.201506290018. We also acknowledge the generous GPU donation from NVIDIA.  ... 
doi:10.1109/tpami.2019.2899857 fatcat:vovtwxufdnawdmwdrb4nepzaam

Learning from Synthetic Data for Opinion-free Blind Image Quality Assessment in the Wild [article]

Zhihua Wang and Zhi-Ri Tang and Jianguo Zhang and Yuming Fang
2021 arXiv   pre-print
Specifically, we first assemble a large number of image pairs from synthetically-distorted images and use a set of full-reference image quality assessment (FR-IQA) models to assign pseudo-binary labels  ...  Here, we propose an opinion-free BIQA method that learns from synthetically-distorted images and multiple agents to assess the perceptual quality of authentically-distorted ones captured in the wild without  ...  ACKNOWLEDGMENT The authors would like to thank Hanwei Zhu from City University of Hong Kong, Xuelin Liu from Jiangxi University of Finance and Economicsfor fruitful discussions throughout the development  ... 
arXiv:2106.14076v3 fatcat:7sbustngfvbxvndavetnshkvve

Entropy Based Data Expansion Method for Blind Image Quality Assessment

Xiaodi Guan, Lijun He, Mengyue Li, Fan Li
2019 Entropy  
Image quality assessment (IQA) is a fundamental technology for image applications that can help correct low-quality images during the capture process.  ...  The ability to expand distorted images and create human visual system (HVS)-aware labels for training is the key to performing IQA tasks using deep neural networks (DNNs), and image quality is highly sensitive  ...  Conflicts of Interest: The authours declare no conflicts of interest.  ... 
doi:10.3390/e22010060 pmid:33285835 pmcid:PMC7516492 fatcat:jyz2c2horfawhhnu2g2wl67ezy

Convolutional Neural Network for Blind Image Quality Assessment

Yadanar Khaing, Yosuke Sugiura, Tetsuya Shimamura
2019 Journal of Signal Processing  
Blind image quality assessment (BIQA) methods can measure the quality of distorted images even without referencing the original images.  ...  This property is indispensable in the image processing field because reference images are normally not available in practice.  ...  As a recent development in BIQA methods, codebook representation for no-reference image quality assessment (CORNIA) [11] , promotes extracting features from the spatial domain, which leads to a sig-nificant  ... 
doi:10.2299/jsp.23.267 fatcat:sbd3y4aclja4ljiff44mphxz5u

A survey on IQA [article]

Lanjiang Wang
2022 arXiv   pre-print
, and focus on the non-reference image quality assessment methods based on deep learning.  ...  Image quality assessment(IQA) is of increasing importance for image-based applications. Its purpose is to establish a model that can replace humans for accurately evaluating image quality.  ...  RankIQA Liu et al. [39] proposed RankIQA , an innovative application of ranked training to learn image quality assessment criteria.  ... 
arXiv:2109.00347v2 fatcat:nbyw6dmyb5a75aecxfbbgh3rem

Joint regression and learning from pairwise rankings for personalized image aesthetic assessment

Jin Zhou, Qing Zhang, Jian-Hao Fan, Wei Sun, Wei-Shi Zheng
2021 Computational Visual Media  
We achieve this in a coarse to fine manner, by joint regression and learning from pairwise rankings.  ...  for the same image.  ...  Acknowledgements The authors thank the reviewers for their valuable comments. This  ... 
doi:10.1007/s41095-021-0207-y fatcat:yve552ozqbahzh66bhkl6tn3lu

Uncertainty-Aware Blind Image Quality Assessment in the Laboratory and Wild [article]

Weixia Zhang and Kede Ma and Guangtao Zhai and Xiaokang Yang
2021 arXiv   pre-print
Extensive experiments on six IQA databases show the promise of the learned method in blindly assessing image quality in the laboratory and wild.  ...  We first sample pairs of images from individual IQA databases, and compute a probability that the first image of each pair is of higher quality.  ...  ACKNOWLEDGEMENTS The authors would like to thank Jingxian Huang for coordinating the subjective experiment and all subjects for participation.  ... 
arXiv:2005.13983v6 fatcat:dkvzaautv5fmvfpsj5c7e54rgu

MSTRIQ: No Reference Image Quality Assessment Based on Swin Transformer with Multi-Stage Fusion [article]

Jing Wang, Haotian Fan, Xiaoxia Hou, Yitian Xu, Tao Li, Xuechao Lu, Lean Fu
2022 arXiv   pre-print
The proposed method outperforms other methods on standard datasets and ranks 2nd in the no-reference track of NTIRE 2022 Perceptual Image Quality Assessment Challenge [53].  ...  Many Image Quality Assessment(IQA) algorithms have been designed to tackle this problem.  ...  TRIQ [29] , MUSIQ [30] are previous works that intergrate ViTs into no-reference image quality assessment.  ... 
arXiv:2205.10101v2 fatcat:teidsj53zjcttmoc43hfqymfzq

A Deep Insight into Measuring Face Image Utility with General and Face-specific Image Quality Metrics [article]

Biying Fu, Cong Chen, Olaf Henniger, Naser Damer
2021 arXiv   pre-print
Our contribution lies in a thorough examination of how different the image quality assessment algorithms relate to the utility for the face recognition task.  ...  The results of image quality assessment algorithms are further compared with those of dedicated face image quality assessment algorithms.  ...  funded by the German Federal Ministry of Education and Research and the Hessian Ministry of Higher Education, Research, Science and the Arts within their joint support of the National Research Center for  ... 
arXiv:2110.11111v2 fatcat:o7jtjzrcbjhr3g6nvfegnlj5dq

Controllable List-wise Ranking for Universal No-reference Image Quality Assessment [article]

Fu-Zhao Ou, Yuan-Gen Wang, Jin Li, Guopu Zhu, Sam Kwong
2020 arXiv   pre-print
No-reference image quality assessment (NR-IQA) has received increasing attention in the IQA community since reference image is not always available.  ...  These degraded images are label-free yet associated with quality ranking information.  ...  Hence, no-reference IQA (NR-IQA) has been the most widely and deepest studied for machine perception [41] , [42] . Commonly, images suffer from two distinct types of distortion.  ... 
arXiv:1911.10566v2 fatcat:6nqryubagfaxrfjpb2e2ta6usy

Residual Networks Based Distortion Classification and Ranking for Laparoscopic Image Quality Assessment

Zohaib Amjad Khan, Azeddine Beghdadi, Mounir Kaaniche, Faouzi Alaya Cheikh
2020 2020 IEEE International Conference on Image Processing (ICIP)  
Unlike existing image quality measures which focus mainly on estimating a quality score, we propose in this paper to formulate the image quality assessment task as a multi-label classification problem  ...  In this context, a crucial step involves an objective assessment of the image quality, which is a two-fold problem requiring both the classification of the distortion type affecting the image and the estimation  ...  Indeed, the results of quality ranking from our method are firstly compared to a recent deep learning based method which employs a Siamese network for ranking before fine-tuning its single trained branch  ... 
doi:10.1109/icip40778.2020.9191111 dblp:conf/icip/KhanBKC20 fatcat:dbswfdk47nh7zb5xi3cddrrshm

CNN-Based Cross-Dataset No-Reference Image Quality Assessment

Dan Yang, Veli-Tapani Peltoketo, Joni-Kristian Kamarainen
2019 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)  
Recent works on no-reference image quality assessment (NR-IQA) have reported good performance for various datasets.  ...  Related work NSS-based NR-IQA methods define the problem as a classification or a regression problem for features that represent natural scene statistics (NSS) or statistics learned from data [45] .  ...  The proposed Siamese architecture for generic (crossdataset) no-reference image quality assessment.  ... 
doi:10.1109/iccvw.2019.00485 dblp:conf/iccvw/YangPK19 fatcat:72ruk7bhmjcjxabnlbjtfcfrna

Hallucinated-IQA: No-Reference Image Quality Assessment via Adversarial Learning

Kwan-Yee Lin, Guanxiang Wang
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
No-reference image quality assessment (NR-IQA) is a fundamental yet challenging task in low-level computer vision community.  ...  Then, we pair the information of hallucinated reference with the distorted image, and forward them to the regressor to learn the perceptual discrepancy with the guidance of an implicit ranking relationship  ...  Related Work No-reference Image Quality Assessment.  ... 
doi:10.1109/cvpr.2018.00083 dblp:conf/cvpr/LinW18 fatcat:kjrwwadfxnba5kkpcuj4lm4ifu

Hallucinated-IQA: No-Reference Image Quality Assessment via Adversarial Learning [article]

Kwan-Yee Lin, Guanxiang Wang
2018 arXiv   pre-print
No-reference image quality assessment (NR-IQA) is a fundamental yet challenging task in low-level computer vision community.  ...  Then, we pair the information of hallucinated reference with the distorted image, and forward them to the regressor to learn the perceptual discrepancy with the guidance of an implicit ranking relationship  ...  Related Work No-reference Image Quality Assessment.  ... 
arXiv:1804.01681v1 fatcat:unxdeyssgbckfhaftq4p4qhx7u
« Previous Showing results 1 — 15 out of 58 results