Compounded Face Image Retrieval Based on Vertical Web Image Retrieval

Text-based image search engine and content-based image retrieval (CBIR) have achieved much progress in commerce and academic community. In addition, a few attempts have been conducted to integrate the two techniques for image retrieval in web context. However, it is very difficult for general search engine based on image context to return highly accurate results. Compound face image retrieval system, based on a novel vertical web image retrieval framework, is designed to make comprehensive uses of text, image content and some related technologies in face detection and recognition. It incorporates high-level semantics and low-level visual features of web image and provides more accurate results and good extensibility for further applications.