Image Transmission Method Based on Prediction Reconstruction and Progressive Remote Rendering

Abstract

In this paper, an efficient remote rendering image transmission method based on the network instability or the excessive number of users in the network is proposed--a pre-computation reconstruction and a progressive rendering image transmission method, which fully utilizes the user's actual operation Continuity and strong correlation between consecutive frames in scene rendering. The user actions are divided into perspective rotation and viewpoint movement. When the perspective rotates, some key frames are selected for transmission. The receiver then uses a series of prior conditions to predict and reconstruct non-key frames based on these key frames; when the viewpoint moves forward, The current frame is predicted using the information of the previous frame, and only the difference between the actual current frame and the predicted frame is transmitted during the transmission, which effectively reduces the amount of data transmission; when the viewpoint is moved backward, the previous frame is used first for the current The central area of the frame is predicted, and then the difference between the actual data and the prediction data of the central area and the new part of the field of view of the image edge are transmitted, thereby greatly reducing the amount of data that needs to be transmitted in the network transmission process.