A novel video summarization technique using weighted combination of color and texture feature
Pooja Tarare, Dipti Jadhav · 2015
Video Summarization acts as an important function in the field of video examination and abstraction. Due to the enormous growth in multimedia techniques, it is becoming increasingly important to be able to navigate the videos efficiently and in shorter time. To address these challenges, we propose a novel Video Summarization technique based on the weighted combination of color and texture features of an image. The color extraction technique uses Color AutoCorrelogram that expresses how the spatial correlation of pairs of colors changes with distance and the texture feature extraction technique uses Color Co-Occurrence Matrix which helps to capture the texture distribution in the image more appropriately. The weighted combination of these features gives better results. The algorithm finds the weighted combination of color and texture feature for making a video summary which can be used across all genres of videos.