Abstract: Video saliency analysis is an important problem in video processing. Traditional video saliency models are often complicated, time consuming, and sensitive to parameters. This paper proposes a novel video saliency computational model based on frequency domain analysis, which is simple and efficient. Firstly, the original video is parsed to a series of consecutive video scene images. Then, the space domain saliency and time domain saliency of the video scene images are computed, respectively. In each domain, the saliency of each frame is estimated by using Renyi entropy. Finally, both space domain and time domain results are fused to obtain the overall saliency. Application in video advertising system and comparison with state-of-the-art approach have demonstrated the effectiveness of the proposed model.

Key words: Video saliency analysis Frequency analysis Renyi entropy Video advertising