A new modeling for visual attention calculation in video coding

To make effective use of video coding bit-steam and support sufficiently high subjective QoE(Quality of Experience), we propose a ROIs (Regions of Interest) detection model based on current achievements of psychology and visual attention. Combined with saliency calculation and eye-movement tracking analysis, the modeling can obtain the ROIs of different viewing duration. According to the ROIs, method of rate adaptation in video is proposed. Through some experiments, we show that our method can improve subjective QoE and bring significant improvement in ROI areas.