Three-dimension shooting and printing system

Abstract

The invention discloses a method and device of three-dimension shooting and printing. The method for the three-dimension shooting and printing comprises a step of acquiring a deep point cloud of a target by utilizing a low-precision plane distance measuring device, a step of fitting the deep point cloud into an initial surface, a step of acquiring a color image of the target by utilizing a color photographic device, a step of processing the color image to acquire a human face area and a background area, a step of detecting facial feature points, a step of fitting the feature points of the facial features into a facial curve, a step of modifying a face curved surface included in the initial surface with the facial curve as a standard to improve the precision of the face curved surface and to obtain a modified face curved surface, a step of mapping a color photo to the modified face curved surface to obtain a color three-dimension model of the photographic target, and a step of outputting the color three-dimension model to a three-dimension printer to obtain a three-dimension printing image. According to the technical scheme, face feature information in the color image is fully utilized, the precision requirement for the plane distance measuring device is reduced, and complete automation of the three-dimension modeling process of three-dimension printing is achieved.

Description

technical field [0001] The invention relates to the technical fields of computer vision, image processing and computer graphics, in particular to a 3D photo printing method and device. Background technique [0002] In the 3D printing industry, establishing a 3D model, that is, forming the printing content, is a key and difficult link. There are two existing 3D modeling methods: (1) Original model: designers use 3D modeling software to create 3D models; (2) reverse engineering: use active scanning to scan existing objects to obtain 3D data, and then use 3D Modeling software is processed into a model similar to an existing object. The 3D modeling method involved in the present invention belongs to the second category. [0003] The existing reverse engineering methods have two disadvantages: (1) 3D scanners are expensive; (2) the modeling process cannot be automated, which consumes a lot of manpower. [0004] Active scanning refers to projecting additional energy to objects,...

AI Technical Summary

Problems solved by technology

[0003] Existing reverse engineering methods have two disadvantages: (1) 3D scanners are expensive; (2) the modeling process cannot be automated and consumes a lot of manpower

However, both PrimeSense or Kinect only provide a bracket-type fixing method to collect depth data changes from a single angle

When it is necessary to collect the full-angle structural information of the object, it can only be moved by placing it on a mobile bracket or a handheld scanning device firmware, which greatly limits the convenience and stability.

[0005] Within the technical scope of the applicant's current understanding, there is no technology similar to the present invention that uses a low-precision planar distance measuring device to solve the problem of 3D photographic printing

Images