Rate-Distortion Optimization for 360-degree Image Considering Visual Attention

Nowadays, 360-degree image is emerging as a new way of experiencing immersive life. It offers users a free viewpoint experience. At the same time, the huge data amount of 360-degree image and video brings the need of efficient coding for this kind of content, in order to achieve real-time transmission, particularly for virtual/augmented reality application. Since only part of the scene is watched by the observer, compressing the whole scene with equal quality is probably inefficient. In this paper, visual attention is considered for 360-degree image coding. By referring to the saliency map of the 360-degree image, the distortion for the rate-distortion optimization (RDO) is modified to ensure better visual experience. The experimental results show that the viewport with higher interest will be rendered with higher quality, and significant bitrate reduction is achieved when the quality measurement is performed in these regions.