An Interactive Depth Map Texture Copy Defect Removal Method

Abstract

The invention relates to an interactive depth image texture copy defect removal method. The patent of the invention uses the depth camera to simultaneously collect the color image and the depth image of the target, establish the corresponding relationship between the color image and the depth image, and adjust the brightness of the color image. Edges (geometric edges, texture edges) are detected and identified (connecting edge fragments with a minimum spanning tree), and texture edges and target geometric edges are classified and marked interactively, and a classification based on spatial neighbors and color map boundaries is constructed. The edge-preserving filter operator of the marked depth map uses the edge marking information on the color image to guide the depth map filtering, so as to achieve the purpose of removing texture copy defects in the depth map while retaining the true geometric boundary of the target surface. The present invention constructs an effective method for the Kinect v2 depth camera based on ToF, which can maintain real geometric features and eliminate texture copy-false geometric boundaries, and the method interaction is simple, convenient and easy to implement.

Description

technical field [0001] The invention belongs to the field of computer graphics, and aims to improve the quality of scene depth maps perceived by consumer-level depth cameras, and in particular relates to an interactive method for eliminating texture copy defects (that is, false object boundaries) in depth maps. Background technique [0002] The depth perception measurement of the target scene is the basis of optical 3D scanning, and it also helps to segment and identify different targets in the scene. The consumer-grade depth camera (Kinect v2 based on ToF distance measurement) can obtain the depth map of the scene in a low-cost and convenient way. However, the depth map corresponds to the 3D surface segment with obvious texture copy defects (caused by color, pattern and material edge) . If the captured depth map is directly used for 3D reconstruction, false geometric features will appear on the fused surface, which will reduce the reconstruction quality of the 3D scanned o...

AI Technical Summary

Problems solved by technology

Due to the strong luminance response of texture materials, patterns, color edges, and target geometric shape edges, although geometric features and texture edges can be effectively detected on RGB images, it is still challenging to effectively and reliably distinguish them.

Images