Institutional Repository of Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
Web Image Mining Based on Modeling Concept-Sensitive Salient Regions | |
Jing Liu; Qingshan Liu; Jinqiao Wang; Hanqing Lu; Songde Ma | |
2006 | |
会议名称 | IEEE International Conference on Multimedia and Expo |
会议录名称 | Proceedings of the 2006 IEEE International Conference on Multimedia and Expo |
会议日期 | July 9-12, 2006 |
会议地点 | Toronto, Ontario, Canada |
摘要 | In this paper, we propose a probabilistic model for Web image mining, which is based on concept-sensitive salient regions without human intervene. Our goal is to achieve a middle-level understanding of image semantics to bridge the semantic gap existing in the field of image mining and retrieval. With the help of a popular search engine, semantically relevant images are collected, and concept-sensitive salient regions are extracted automatically based on an attention model. Then the semantic concept model is learned from the joint distribution of all salient regions with Gaussian mixture model and expectation-maximization algorithm. In addition, by incorporating semantically irrelevant un-salient regions as negative samples, the discriminative power of the solution is further enhanced. Experiments demonstrate the encouraging performance of the proposed method |
关键词 | Gaussian Processes Data Mining Expectation-maximisation Algorithm Image Retrieval Search Engines Semantic Web Gaussian Mixture Model Web Image Mining Concept-sensitive Salient Region Model Exoexpectation-maximization Algorithm Image Retrieval Image Semantics Probabilistic Model Search Engine Bridges Detectors Html Humans Image Analysis Image Retrieval Image Segmentation Pixel Search Engines Skin |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/13455 |
专题 | 模式识别国家重点实验室_图像与视频分析 |
通讯作者 | Jing Liu |
推荐引用方式 GB/T 7714 | Jing Liu,Qingshan Liu,Jinqiao Wang,et al. Web Image Mining Based on Modeling Concept-Sensitive Salient Regions[C],2006. |
条目包含的文件 | 条目无相关文件。 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论