Perceptual high-definition video coding method based on salient target detection and saliency guidance

Abstract

The invention discloses a perceptual high-definition video coding method based on salient target detection and saliency guidance. The method comprises the following steps: constructing a salient target detection model of a multi-scale pyramid shuffling network; carrying out salient region prediction on video data through the salient target detection model of the multi-scale pyramid shuffling network; and guiding an HEVC video compression standard by utilizing a prediction result, and performing video coding through an adaptive quantization parameter and a significance-based coding unit partitioning strategy. The significant target detection model of the multi-scale pyramid shuffling network is stronger in generalization, and can output a prediction result image of significant target segmentation with higher accuracy; the HEVC video compression standard is guided on the basis of the prediction result image, the video image is divided into a salient region and a non-salient region, dynamic optimization is carried out on rate distortion optimization and quantization parameter selection, finally, a video coding result with better indexes is obtained, the video code stream is smaller, and the image quality is better.

Description

technical field [0001] The invention relates to the field of video technology, in particular to a perceptual high-definition video coding method based on salient object detection and saliency guidance. Background technique [0002] In the information age, with the rapid development of video technology and applications, visual information carriers such as video and images have wider practicability and higher use efficiency, giving full play to their intuition, certainty, efficiency and The high-bandwidth characteristics of video signals have penetrated into all aspects of our work and life. [0003] At present, the videos that people watch on various channels and devices are all compressed videos. If there is no step of video compression, the original image quality and original code stream video will have a considerable amount of data, which is unacceptable for data transmission. At present, the fastest transmission medium optical fiber can only reach 100Mbps. Compression, ...

AI Technical Summary

Problems solved by technology

[0005] In addition, the existing state-of-the-art Salient Object Detection Networks (salient object detection network) are based on pre-trained convolutional neural networks (CNNs) on massive datasets, and they are not very good at multi-scale features in CNN models. It is fully utilized and integrated, and the prediction results of Salient Object Detection (salient object detection) are not well used in engineering applications such as video processing. At the same time, in terms of video compression, the most advanced high-efficiency video coding (HEVC) in video compression There is still room for improvement in the size and image quality of the subsequent code stream

Images