Image blur classification and blur usefulness assessment

Fan, Mingyuan; Huang, Rui; Feng, Wei; Sun, Jizhou

doi:10.1109/icmew.2017.8026291

Search citation statements

Order By: Relevance

Paper Sections

Select...

Attributes As Privileged Information In Photo Aesthetics1

Photo-based Attributes1

Citation Types

Supporting

Mentioning

Contrasting

Year Published

2020

2024

Publication Types

Select...

Article3

Other2

Relationship

Self Cite0

Independent5

Authors

Journals

Cited by 5 publications

(2 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The aesthetic attributes of photos are usually exploited from the perspective of photo and photography. Attributes from photo mean the robust feature representations describing the aesthetic aspect of a photo, such as color [21], content [29], light [30], depth of field [29], rule of third [29] and motion blurs [31] etc. Such attributes are related to the artistic aspect of photos and can evaluate the quality of photos.…”

Section: Attributes As Privileged Information In Photo Aestheticsmentioning

confidence: 99%

See 1 more Smart Citation

Learning with privileged information for photo aesthetic assessment

Shu

Liu

et al. 2020

Neurocomputing

View full text Add to dashboard Cite

Section: Attributes As Privileged Information In Photo Aestheticsmentioning

confidence: 99%

“…It occurs when the photo is recorded changes during the recording due to rapid movement or long exposure. Thus, under moving objects in a photograph condition, a photo is more likely to invoke high aesthetic quality of photo when it contains motion blur [31].…”

Section: Photo-based Attributesmentioning

confidence: 99%

Learning with privileged information for photo aesthetic assessment

Shu

Liu

et al. 2020

Neurocomputing

View full text Add to dashboard Cite

Semi-Supervised Ranking for Object Image Blur Assessment

Yao

Wang

et al. 2022

2022 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

Assessing the blurriness of an object image is fundamentally important to improve the performance for object recognition and retrieval. The main challenge lies in the lack of abundant images with reliable labels and effective learning strategies. Current datasets are labeled with limited and confused quality levels. To overcome this limitation, we propose to label the rank relationships between pairwise images rather their quality levels, since it is much easier for humans to label, and establish a large-scale realistic face image blur assessment dataset with reliable labels. Based on this dataset, we propose a method to obtain the blur scores only with the pairwise rank labels as supervision. Moreover, to further improve the performance, we propose a self-supervised method based on quadruplet ranking consistency to leverage the unlabeled data more effectively. The supervised and self-supervised methods constitute a final semi-supervised learning framework, which can be trained end-to-end. Experimental results demonstrate the effectiveness of our method. Source of labeled datasets: https://github.com/yzliangHIK2022/SSRanking-for-Object-BA

show abstract