A Survey On Text Based Video Retrieval Using Semantic And Visual Approach
In recent past, the cost for storing multimedia data and the digital multimedia storage grows is cheaper. So there is
abundant number of videos available in the video repositories. Video sharing on the web is growing with a tremendous speed
which creates perhaps the most heterogeneous and the largest publicly available video archive. There are vast amount of video
archives including broadcast news, documentary videos, meeting videos, movies etc. Therefore video retrieval i.e, searching
and retrieving videos relevant to a user defined query is one of the most popular topics in both multimedia research and real
life applications. Text data present in the multimedia contains useful information for retrieving and indexing purpose. In
current video search, the search results are influenced by the metadata information such as title, captions associated with
them. Extracted text can be used to recognize the overlay, scene text, which is used for video retrieval. The main focus of work
is 1) text retrieval and 2) semantic retrieval. The semantic concept: an objective linguistic description of an observable entity.
The main contribution of this proposal is utilization of the semantic word similarity measures for the text-based retrieval of
videos. This paper offers an overview of the different existing techniques in text , image and video based video retrieval and
different approaches to search with in long videos .
Keywords- CBIR, Key Frames, Multimedia Retrieval, Similarity Matching Algorithm, Text Frames Classification, Video
Indexing, Video Retrieval System.