Flexible photo retrieval (FlexPhoReS) : a prototype for multimodel personal digital photo retrieval

Digital photo technology is developing rapidly and is motivating more people to have large personal collections of digital photos. However, effective and fast retrieval of digital photos is not always easy, especially when the collections grow into thousands. World Wide Web (WWW) is one of the platforms that allows digital photo users to publish a collection of photos in a centralised and organised way. Users typically find their photos by searching or browsing uSing a keyboard and mouse. Also in development at the moment are alternative user interfaces such as graphical user interfaces with speech (S/GUI) and other multimodal user interfaces which offer more flexibility to users. The aim of this research was to design and evaluate a flexible user interface for a web based personal digital photo retrieval system. A model of a flexible photo retrieval system (FlexPhoReS) was developed based on a review of the literature and a small-scale user study. A prototype, based on the model, was built using MATLAB and WWW technology. FlexPhoReS is a web based personal digital photo retrieval prototype that enables digital photo users to . accomplish photo retrieval tasks (browsing, keyword and visual example searching (CBI)) using either mouse and keyboard input modalities or mouse and speech input modalities. An evaluation with 20 digital photo users was conducted using usability testing methods. The result showed that there was a significant difference in search performance between using mouse and keyboard input modalities and using mouse and speech input modalities. On average, the reduction in search performance time due to using mouse and speech input modalities was 37.31%. Participants were also significantly more satisfied with mouse and speech input modalities than with mouse and keyboard input modalities although they felt that both were complementary. This research demonstrated that the prototype was successful in providing a flexible model of the photo retrieval process by offering alternative input modalities through a multimodal user interface in the World Wide Web environment.