Hybrid image generation method and device

Abstract

The invention provides a hybrid image generation method and device, and the method comprises the steps: building a post-processing material, and determining a core radius according to a screen space corresponding to the post-processing material; determining a current sampling point coordinate in the screen space, and obtaining a corresponding deviated sampling point coordinate according to the current sampling point coordinate, the kernel radius and a deviation parameter; sampling the original background image in the screen space according to the deviated sampling point coordinates to obtain acorresponding deviated color sample; and obtaining a final mixed color corresponding to the current sampling point according to the plurality of deviated color samples, and generating a mixed image in the screen space according to the final mixed color corresponding to each sampling point. The final mixed color corresponding to each sampling point is generated by fusing a plurality of shifted color samples, so that the mixed color in the finally generated mixed image has continuity, and an obvious boundary can be eliminated.

Description

technical field [0001] The present application relates to the technical field of image processing, and in particular to a method and device for generating a mixed image, a computing device, and a computer-readable storage medium. Background technique [0002] In the field of image processing, image blending is often used. In a large-screen project, the 3D background rendering layer is a city with green mountains and green waters. The 2d layer is mixed with the 3d layer after the UI (User Interface, user interface) area and icons are confirmed. At this time, the background of the UI area needs to be blurred to present the texture of frosted glass. [0003] In the prior art, the following methods are generally adopted: [0004] Transparent channel mixing, that is, directly mixing the 3d background layer and the 2d UI layer according to the transparency. The disadvantage of this method is that it lacks the texture of frosted glass and cannot meet the design requirements. [...

AI Technical Summary

Problems solved by technology

[0004] Transparent channel mixing, that is, directly mixing the 3D background layer and the 2D UI layer according to the transparency. The disadvantage of this method is that it lacks the texture of frosted glass and cannot meet the design requirements.

[0005] Template-based Gaussian fuzzy mixing, that is, directly using binary templates for Gaussian fuzzy mixing, the disadvantage of this method is that the fuzzy mixing lacks continuity and has obvious boundaries

Images