VGRAPH: An Effective Approach for Generating Static Video Summaries

A video summary is a sequence of still pictures that represent the content of a video in such a way that the respective target group is rapidly provided with concise information about the content, while the essential message of the original video is preserved. In this paper, we present VGRAPH, a simple yet effective video summarization approach that utilizes both color and texture features. This approach is based on partitioning the video into shots by utilizing the color features, and extracting video key frames using a nearest neighbor graph built from the texture features of the shots representative frames. Also, this paper introduces and illustrates an enhanced evaluation method based on color and texture matching. Video summaries generated by VGRAPH are compared with summaries generated by others found in the literature and the ground truth summaries. Experimental results indicate that the video summaries generated by VGRAPH have a higher quality than others.