Delhi
Multimedia Systems and Computer VisionChallengesUbiquity of image, video and audio capture devices have resulted in large volumes of data being recorded and stored in multimedia form. Today, multimedia collections dominate the information space, as compared to traditional databases and text-bases in the last decades. While text and numbers are creations of human beings, they are accustomed to deal with multimedia data more “naturally” and intuitively. Technological limitations restrict existing most of the rich-media applications to acquisition, storage, conversion, transmission and rendering of media data without interpreting their information content. Few applications that attempts to interpret media data are restricted to specific media forms and specific domains. OverviewThe goal of Multimedia Technology group at the TCS Innovation Labs-Delhi is to create technologies for build modern-age information systems that can deal with multimedia objects, reason with media contents and can effectively interact with human beings using multimedia artifacts. It seeks to create a framework for a Semantic Multimedia Web that can enable machine-based interpretation and processing of different media forms distributed on the web. Research activities of the team are geared towards creating prototype solutions that demonstrate capabilities of such technology towards effective interaction with human beings in pragmatic application scenario Some of our key research areas are:
Major Research ProjectsBroadcast Analytics: Integrated News Analysis Analysis of public newscast by domestic as well as foreign TV channels for tracking news, national and international views and public opinion, is of paramount importance for media analysts in several domains, such as journalism, brand monitoring, law enforcement and internal security. The channels representing different countries, political groups, religious conglomerations and business interests present different perspectives and viewpoints of the same event. The motivation of this research project is to automate processing of news video streams in multiple languages. While there has been significant research on automated processing of news video, processing transmissions in Indian (and some other) languages poses additional challenges because of unavailability of reliable language tools. This research projects aims at exploring multimodal analysis techniques using audio-visual cues for processing news transmissions. Towards a Virtual University Digital Education brings education out of brick-and-mortar classroom and uses electronic medium to cut across the geographical boundaries. It has a huge potential in India as well as in other countries to bring quality education to the desktops of millions of students and professionals. We envisage that the tele-teaching system will be situated in a Digital University environment, which comprise an eco-system of many stake-holders. While the faculty and the students will be actively engaged in education, the supporting activities like content creation, administration and advancement of learning will be facilitated by instructional designers, content creators and subject experts. Multimedia contents will play a progressively important role in educational material. This research project is aimed at exploring tools and technologies for creation, storage, retrieval, playback and reuse of multimedia educational material. Another goal of the project is to create a classroom-like teacher-student interactivity despite possible geographical and temporal separation. In computer vision the prime focus is on developing a multi-camera surveillance system and doing analytics on top of it. EventsAlliancesSolutions
|
Google
Facebook
Twitter
del.icio.us
Blogger
Rain Concert
TCS is a member of the Tata Group 