Multimodal Search on Mobile Devices

Xin Fan,Mark Sanderson,Xing Xie
DOI: https://doi.org/10.4018/978-1-60566-978-6.ch010
2010-01-01
Abstract:The increasingly popularity of powerful mobile devices, such as smart phones and PDAs, enables users to search for information on the move. However, text is still the main input modality in most current mobile search services although some providers are attempting to provide voice-based mobile search solutions. In this chapter, we explore the innovative query modalities to enable mobile devices to support queries such as text, voice, image, location, and their combinations. We propose a solution to support mobile users to perform visual queries. The queries by captured pictures and text information are studied in depth. For example, the user can simply take a photo of an unfamiliar flower or surrounding buildings to find related information from the Web. A set of indexing schemes are designed to achieve accurate results and fast search through large volumes of data. Experimental results show that our prototype system achieved satisfactory performance. Also, we briefly introduce a prospective mobile search solution based on our ongoing research, which supports multimodal queries including location information, captured pictures and text information.
What problem does this paper attempt to address?