SALI360 provides a new video encoding algorithm for efficient 360 video streaming service by understanding Human Vision System (HVS) and human behavior. Unique characteristics of HVS, of which visual perception ability precipitously falls along with its distance to visual focus, indicates a huge potential to efficiently reduce bandwidth without notice in quality variance in content during streaming service. SALI360 utilizes saliency and user behavior prediction algorithms to predict possible focal areas and pre-encodes those areas in higher resolution according to network conditions, thereby improving user perception quality.

In order to evaluate perceived quality, we define a new novel Quality of Perception (QoP) measurement that measures visual quality in Field of Look (FoL) region instead of traditional Field of View (FoV). Results of our experimental 360 VR video streaming platform show that SALI360 achieves 53.3% of PSNR improvement in FoL while gaining 9.3% of PSNR improvement in FoV. In addition, our subjective study on 93 participants verifies that SALI360 improves viewers’ QoP in the 360 VR streaming service.