Analysis of Video Shot Detection using Color Layout Descriptor and Video Summarization based on Expectation-Maximization Clustering

Video Shot Detection is used to determine the scene changes in the video. Video Summarization provides most informative frames to generate the abstract of the input video. In this paper, we have integrated both the concepts for the better outcome of the results. The Color Layout Descriptor and Expected-Maximization are used for the Video Shot Detection and Video summarization respectively. The output frames of Color Layout Descriptor are used as input for the Expected-Maximization Clustering method which provides the final summarized frames. The quality metric parameters such as Recall and Precision are used to evaluate the robustness of the key frame extraction strategy. Experimental results show that the proposed method provides better results compare to the existing methodologies.