Linear shadow mapping method of de-pixeldined contour line reconstruction

Abstract

The invention discloses a linear shadow mapping method of de-pixeldined contour line reconstruction. The linear shadow mapping method comprises the following steps: (1) drawing a scene from a light source space, storing a depth value of a pixel center as a first shadow mapping, deviating in x and y directions for half texture pixel to obtain a second shadow mapping, and forming a linear shadow mapping by the first shadow mapping and the second shadow mapping; (2) drawing the scene from a vision field space, converting pixels into the light source space, and classifying the pixels into contour line pixels and non-counter line pixels; and (3) for the non-counter line pixels, using linear interpolation to calculate the depth of an occluder, and comparing the depth value to obtain a shadow result, and reconstructing a shadow boundary by a de-pixeldined contour line reconstruction method for the contour line pixels. The linear shadow mapping method of the de-pixeldined contour line reconstruction can guarantee the smoothness of the shadow counter lines and can effectively weaken a deformation phenomenon in the shadow mapping method by small storage overheads.

Description

technical field [0001] The invention relates to the field of shadow drawing in computer graphics, in particular to a linear shadow map method for reconstructing depixelated contour lines. Background technique [0002] Shadows are one of the important elements of the real world and an integral part of photorealistic rendering of composite images. It carries important information such as the position of the light source in the scene and the direction of light transmission. For any visible point on the screen, it is either directly illuminated by the light source and is in the illuminated area, or it is blocked by other objects and is in the shadow area. For users, correct shadow information can help users effectively understand the geometric properties of objects and distinguish the relative positional relationship between objects. [0003] Although shadow drawing algorithms have been a classic problem in computer graphics, the problem of how to efficiently and robustly gene...

AI Technical Summary

Problems solved by technology

However, the shadow map algorithm also faces two major challenges: aliasing, which is caused by the limited resolution of the shadow map and insufficient sampling of depth information; self-shadowing, which usually manifests itself as the originally illuminated objects in the scene Showing incorrect shadow stripes, as if the object is occluded by itself, there are two reasons for this - numerical quantization error and discrete sampling of depth cause the depth value of the shadow map to be accurate only in the center of the texel

[0004] Improving the resolution of the shadow map is the most direct way to solve the inherent problems of the shadow map, but too large a shadow map resolution will bring serious storage and performance overhead to the algorithm

Images