Perceptual Video Coding Based on Spatial Masking for Medical Video Communication

In recent years, medical video compression has been developing rapidly in medical video communication applications for remote diagnosis and emergency medical incidents. How to improve the diagnostically perceptual quality with limited bandwidth in low-delay requirement has become a major obstacle limiting the development of medical video communication. In order to solve the above problems, we propose an adaptive quantization parameter algorithm based on fast spatial masking estimation to reduce bitrate while maintain near lossless perceptual quality. Since the computation of spatial masking is time-consuming, we propose a spatial masking estimation method based on dual support vector regression model to meet the real-time requirement. Experimental results show that the proposed method achieves more than 10% bitrate reduction with unchanged perceptual quality, while meets real-time requirements.