The Design of High-Level Features for Photo Quality Assessment
Y. Ke, X. Tang, und F. Jing. Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 1, Seite 419--426. Washington, DC, USA, IEEE Computer Society, (2006)
DOI: 10.1109/CVPR.2006.303
Zusammenfassung
We propose a principled method for designing high level features forphoto quality assessment. Our resulting system can classify between high quality professional photos and low quality snapshots. Instead of using the bag of low-level features approach, we first determine the perceptual factors that distinguish between professional photos and snapshots. Then, we design high level semantic features to measure the perceptual differences. We test our features on a large and diverse dataset and our system is able to achieve a classification rate of 72% on this difficult task. Since our system is able to achieve a precision of over 90% in low recall scenarios, we show excellent results in a web image search application.
Beschreibung
The Design of High-Level Features for Photo Quality Assessment
%0 Conference Paper
%1 Ke:2006:DHF:1153170.1153495
%A Ke, Yan
%A Tang, Xiaoou
%A Jing, Feng
%B Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 1
%C Washington, DC, USA
%D 2006
%I IEEE Computer Society
%K comp_photography photography
%P 419--426
%R 10.1109/CVPR.2006.303
%T The Design of High-Level Features for Photo Quality Assessment
%U http://dx.doi.org/10.1109/CVPR.2006.303
%X We propose a principled method for designing high level features forphoto quality assessment. Our resulting system can classify between high quality professional photos and low quality snapshots. Instead of using the bag of low-level features approach, we first determine the perceptual factors that distinguish between professional photos and snapshots. Then, we design high level semantic features to measure the perceptual differences. We test our features on a large and diverse dataset and our system is able to achieve a classification rate of 72% on this difficult task. Since our system is able to achieve a precision of over 90% in low recall scenarios, we show excellent results in a web image search application.
%@ 0-7695-2597-0
@inproceedings{Ke:2006:DHF:1153170.1153495,
abstract = {We propose a principled method for designing high level features forphoto quality assessment. Our resulting system can classify between high quality professional photos and low quality snapshots. Instead of using the bag of low-level features approach, we first determine the perceptual factors that distinguish between professional photos and snapshots. Then, we design high level semantic features to measure the perceptual differences. We test our features on a large and diverse dataset and our system is able to achieve a classification rate of 72% on this difficult task. Since our system is able to achieve a precision of over 90% in low recall scenarios, we show excellent results in a web image search application.},
acmid = {1153495},
added-at = {2014-09-12T14:17:52.000+0200},
address = {Washington, DC, USA},
author = {Ke, Yan and Tang, Xiaoou and Jing, Feng},
biburl = {https://www.bibsonomy.org/bibtex/2d25962464b2703dde6fe73edde012438/alex_ruff},
booktitle = {Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 1},
description = {The Design of High-Level Features for Photo Quality Assessment},
doi = {10.1109/CVPR.2006.303},
interhash = {91ff0f33aa513eb48835e0eb894e86c8},
intrahash = {d25962464b2703dde6fe73edde012438},
isbn = {0-7695-2597-0},
keywords = {comp_photography photography},
numpages = {8},
pages = {419--426},
publisher = {IEEE Computer Society},
series = {CVPR '06},
timestamp = {2014-09-12T14:17:52.000+0200},
title = {The Design of High-Level Features for Photo Quality Assessment},
url = {http://dx.doi.org/10.1109/CVPR.2006.303},
year = 2006
}