|
使用机器学习评价摄影美学
Evaluating Photo Aesthetics Using Machine Learning |
|
| |
| 课程网址: |
http://videolectures.net/is2012_pogacnik_photo_aesthetics/
|
| 主讲教师: |
Domen Pogačnik
|
| 开课单位: |
卢布尔雅那大学
|
| 开课时间: |
2012-11-16
|
| 课程语种: |
英语
|
| 中文简介: |
在本文中，我们提出了一种自动评估照片审美吸引力的方法。我们确定了将高质量摄影与低质量快照区分开来的重要参数。在这些参数的基础上，我们使用机器学习方法定义了可自动评估摄影美学的可计算特征。特征的计算在很大程度上取决于照片中对象的识别。随着主题的识别，我们定义并实现了各种功能来分析照片的各个方面。在两个数据集上测试了这些功能。从Flickr获得第一个数据集并手动标记以进行评估。第二数据集基于来自DPChallenge门户的照片，其中受试者用面部检测算法识别。两项实验均显示出一些有希望在本文中，我们指定了有助于成功分类照片，分析其影响并讨论结果的特征。总之，我们提出了一些进一步研究的建议。
|
| 课程简介: |
In this paper we propose a method for automatic assessment of aesthetic appeal of photographs. We identify signifi cant parameters that distinguish high quality photography from low quality snapshots. On the basis of these parameters, we de fined calculable features for automatic assessment of photography aesthetics using machine learning methods. The calculation of features depends heavily on the identifi cation of the subject in photographs. With the subject identi fied, we defi ned and implemented various features to analyze various aspects of a photograph. The features were tested on two datasets. First dataset was obtained from Flickr and manually labeled for evaluation. Second dataset was based on photographs from DPChallenge portal where subjects were identi ed with a face detection algorithm. Both experiments showed some promising results. In this article we specify the features which contribute to a successful classi cation of photographs, analyze their in influence and discuss the results. In conclusion, we off er some suggestions for further research.
|
| 关 键 词: |
重要参数; 自动评估; 功能测试
|
| 课程来源: |
视频讲座网
|
| 最后编审: |
2020-06-27：zyk
|
| 阅读次数: |
48
|
|
图 书 馆