Parallel Computing in Face Image Retrieval

Modern digital photo collections contain vast multitudes of high-resolution color images, many containing faces, which are desirable to retrieve visually. This poses a problem for effective image browsing and calls for efficient Content Based Image Retrieval (CBIR) capabilities ensuring near-instantaneous visual query turn-around. This in turn necessitates parallelization of many existing image processing and information retrieval algorithms that can no longer satisfy the modern user demands, when executed sequentially. Hence a practical approach to Face Image Retrieval (FIR) is presented. It utilizes multi-core processing architectures to implement its major modules (e.g. face detection and matching) efficiently without sacrificing the image retrieval accuracy. The integration of FIR into a web-based family reunification system demonstrates the practicality of the proposed method. Several accuracy and speed evaluations on real-word data are presented and possible CBIR extensions are discussed.

Download Full-text

Face Match for Family Reunification

Computer Vision ◽

10.4018/978-1-5225-5204-8.ch028 ◽

2018 ◽

pp. 735-753

Author(s):

Eugene Borovikov ◽

Szilard Vajda ◽

Michael Gill

Keyword(s):

Face Detection ◽

Family Reunification ◽

Face Image ◽

Digital Photo ◽

Photo Collections ◽

The Many ◽

Fir System ◽

Post Disaster ◽

Person Location ◽

Variable Quality

Despite the many advances in face recognition technology, practical face detection and matching for unconstrained images remain challenging. A real-world Face Image Retrieval (FIR) system is described in this paper. It is based on optimally weighted image descriptor ensemble utilized in single-image-per-person (SIPP) approach that works with large unconstrained digital photo collections. The described visual search can be deployed in many applications, e.g. person location in post-disaster scenarios, helping families reunite quicker. It provides efficient means for face detection, matching and annotation, working with images of variable quality, requiring no time-consuming training, yet showing commercial performance levels.

Download Full-text

WEB-BASED PERSONAL DIGITAL PHOTO COLLECTIONS: MULTIMODAL RETRIEVAL

IIUM Engineering Journal ◽

10.31436/iiumej.v10i1.104 ◽

2010 ◽

Vol 10 (1) ◽

pp. 49-57

Author(s):

Nor Azman Ismail ◽

Ann O'Brien

Keyword(s):

User Study ◽

Multimodal Interaction ◽

Prototype System ◽

Web Based ◽

Digital Photo ◽

Photo Collections ◽

It Architecture ◽

Multimodal Retrieval ◽

Data Files ◽

Retrieval Engine

When personal photo collections get large retrieval of specific photos or sets of photos becomes difficult mainly due to the fairly primitive means by which they are organised. Commercial photo handling systems help but often have only elementary searching features. In this paper, we describe an interactive web-based photo retrieval system that enables personal digital photo users to accomplish photo browsing by using multimodal interaction. This system not only enables users to use mouse click input modalities but also speech input modality to browse their personal digital photos in the World Wide Web (WWW) environment. The prototype system and it architecture utilise web technology which was built using web programming scripting (JavaScript, XHTML, ASP, XML based mark-up language) and image database in order to achieve its objective. All prototype programs and data files including the user’s photo repository, profiles, dialogues, grammars, prompt, and retrieval engine are stored and located in the web server. Our approach also consists of human-computer speech dialogue based on photo browsing of image content by four main categories (Who? What? When? and Where?). Our user study with 20 digital photo users showed that the participants reacted positively to their experience with the system interactions.

Download Full-text