Three-dimensional reconstruction method combined with image main body extraction technology

• Main Authors: CHENG XIUCHAO, WU LEI, YE XUCHAO, YIN ZHICHENG
• Format: Patent
• Language: Chinese; English
• Published: 05.09.2023
• Subjects: CALCULATING; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

The invention relates to a three-dimensional reconstruction method combined with an image main body extraction technology, and the method comprises the steps: sequentially placing foreground objects into at least two different spatial postures, and collecting image sequences respectively; performing interactive image segmentation on part of the image, generating a positive label by clicking a foreground, generating a negative label by clicking a background, and generating a segmentation result; in combination with image feature extraction and an image feature matching algorithm, the positive label and the negative label are sequentially transmitted to the image pair with the highest matching degree through image feature matching points, and therefore image segmentation is completed; taking the image after main body extraction and an image matching result as input, executing an SfM process, completing camera parameter calculation, and triangularizing sparse point clouds according to the calculated camera param