Depth map post-processing method

Abstract

The invention relates to an image post-processing method. The invention is to solve the technical problem that the depth map post-processing method capable of effectively improving the accuracy of depth map information is provided. The technical scheme adopted in the invention is characterized in that the depth map post-processing method comprises the following steps: to begin with, obtaining a current frame image and a previous frame image in an original two-dimension video; utilizing optical flow motion to detect the current frame image and the previous frame image to obtain a motion vector of the current frame image; judging whether scene change occurs in the current frame image and the previous frame image, if so, performing smooth filtering to obtain a new depth map of the current frame image, and if not, obtaining motion parameters of the current frame image and an existing depth map of the previous frame image and obtaining a virtual depth map of the current frame image; performing shield detection on the virtual depth map and an existing depth map of the current frame image; and performing smooth filtering to obtain a new depth map of the current frame image. The depth map post-processing method is suitable for carrying out post-processing on the depth map.

Description

technical field [0001] The invention relates to an image post-processing method, in particular to a depth map post-processing method. Background technique [0002] The generation of image depth maps has been a major research direction in computer vision. Depth map information can help people recalculate and synthesize two-dimensional image objects into a three-dimensional coordinate system with multiple viewpoints, so as to objectively express the three-dimensional effect of objects. [0003] Depth map information is widely used in computer-aided geometric design (CAGD), computer graphics (CG), computer animation, computer vision, medical image processing, scientific computing, and virtual reality. At present, the commonly used depth map generation method mainly uses the geometric information of the planar video, camera parameters, color features of objects in the planar image, etc. to obtain locally or globally. However, compared with the color image, the texture of the d...

AI Technical Summary

Problems solved by technology

However, due to the limitations of the depth map information generation method, there are many problems in the depth map image, such as poor time continuity, discontinuous depth information, inaccurate contour information, etc.

What's more, the depth map information should not be directly applied to the front end, but used to assist 3D reconstruction, virtual view drawing, etc., so inaccurate depth information will greatly affect its stereo effect

Images