Dominant frame extraction for video indexing
N Shruthi, S Priyamvada · 2017
With the Increase in the multimedia technology a large amount of video information is collected. Due to time constraint, it is not possible for viewer to watch the entire video. This problem is overcome by using video indexing methods. In this paper, we explain the robust video indexing techniques by using different features extraction model with dominant frame generation for each frame of the input video. The overall system model is divided into two phases: training and testing. Using dominant frames the important seen or events of the video is captured, and the dominant frames are then used in video indexing. SVM classifier is used to classify the set of frames into dominant frames. CLBP and Tensor LPP feature extraction algorithm is used to extract the texture features from the video frame. The performance of the designed model is evaluated and the discussion about the same is presented in result section.