Nilesh Bhojne¹, Pravinkumar Kamde¹ and S. P. Algur², ¹Sinhgad College of Engineering, India and ²RC University Belguam, India

From the last decade the Multi/Rich Media information production and usage is growing extremely. Video is the most informative and challenging as it is a combination of all other media. The way by which the video is present for access has become a challenging task both for system and the viewer applications. At all stages and for all target groups, an effective and efficient video retrieval facility is becoming necessary. In current video search, the search results are influenced by the metadata information such as title, captions associated with them. Our main focus of work is the notion of a semantic concept: an objective linguistic description of an observable entity. This can be very well achieved by labeling combinations of anchor person, persons and objects appearing in news, events appearing in the audiovisual content. News captions are most often found to contain information about the news being telecasted. In this paper, we present news video retrieval solution that target specific news videos based on their contents described by overlay text. The proposed approach is based on use of overlay text that conveys direct meaning of video as a source of complementary information. The whole process is divided in to two steps. Firstly, we build the “metadata labels” by detecting and extracting the overlay text. Secondly, these labels are then used to index the news videos. The experiments are carried on the news videos from NDTV News and large data set of video images containing artificial text developed at Image Processing Center (IPC) a research facility at - National University of Sciences and Technology (NUST), Pakistan..

News Video, Overlay Text, Video OCR, Transition map