Association and Content-Based Retrieval
IEEE Transactions on Knowledge and Data Engineering
Latent semantic analysis of facial action codes for automatic facial expression recognition
Proceedings of the 6th ACM SIGMM international workshop on Multimedia information retrieval
Data Mining: Concepts and Techniques
Data Mining: Concepts and Techniques
An interactive system for mental face retrieval
Proceedings of the 7th ACM SIGMM international workshop on Multimedia information retrieval
A survey of content-based image retrieval with high-level semantics
Pattern Recognition
Supervised Learning of Semantic Classes for Image Annotation and Retrieval
IEEE Transactions on Pattern Analysis and Machine Intelligence
Modeling Semantic Aspects for Cross-Media Image Indexing
IEEE Transactions on Pattern Analysis and Machine Intelligence
Image retrieval: Ideas, influences, and trends of the new age
ACM Computing Surveys (CSUR)
Introduction to Information Retrieval
Introduction to Information Retrieval
A survey of methods for image annotation
Journal of Visual Languages and Computing
Annotating Images by Mining Image Search Results
IEEE Transactions on Pattern Analysis and Machine Intelligence
Multimedia Data Mining: A Systematic Introduction to Concepts and Theory
Multimedia Data Mining: A Systematic Introduction to Concepts and Theory
Face Image Annotation in Impressive Words by Integrating Latent Semantic Spaces and Rules
KES '09 Proceedings of the 13th International Conference on Knowledge-Based and Intelligent Information and Engineering Systems: Part II
Active learning in multimedia annotation and retrieval: A survey
ACM Transactions on Intelligent Systems and Technology (TIST)
| Hi-index | 0.00 |
This paper describes a system to annotate and to retrieve face images in impressive words representing their visual impressions. When a face image is given, impressive words are assigned by annotation. When some impressive words are given, face images are obtained by retrieval. In order to achieve them, latent semantic spaces, association rules and decision trees are utilized, which are constructed from a set of face image descriptions. The face image is described in visual and symbolic features. Visual features are sizes and/or lengths of the face parts, symbolic features are impressive words, respectively. Two types of visual feature are defined, which are 24 places and minimum bounding rectangles. In the former, the lengths of 24 places in a face are measured. In the latter, minimum bounding rectangles of the face parts are made, and lengths between the rectangles are measured. Efficiency of annotation and retrieval are evaluated using these two types of visual feature. Experimental results using minimum bounding rectangles are better than ones using 24 places in both annotation and retrieval.