High-frequency region enhanced photometric three-dimensional reconstruction method based on deep learning

Abstract

A high-frequency region enhanced photometric three-dimensional reconstruction method based on deep learning comprises the steps of shooting a plurality of images of an object to be reconstructed by using a photometric three-dimensional system, and outputting accurate surface normal three-dimensional reconstruction by using a deep learning algorithm, designing a surface normal generation network to generate a surface normal of an object needing to be reconstructed from an image and illumination; enabling the attention weight generation network to generate an attention weight map of an object needing to be reconstructed from the image; processing the attention weight loss function pixel by pixel; and using the trained network to carry out surface normal reconstruction of the photometric stereo image. According to the method, the surface normal information and the high-frequency information are learned respectively through the proposed surface normal generation network and the attention weight generation network, training is carried out by utilizing the proposed attention weight loss, and the reconstruction precision of the surface of the high-frequency region such as the wrinkle edge can be improved. Compared with a previous traditional photometric stereo method, the method improves the three-dimensional reconstruction precision, especially the details of the surface of the object to be reconstructed.

Description

technical field [0001] The invention relates to a high-frequency area enhanced photometric three-dimensional reconstruction method based on deep learning, which belongs to the field of multi-degree three-dimensional reconstruction. Background technique [0002] The 3D reconstruction algorithm is a very important and basic problem in computer vision. The photometric stereo algorithm is a high-precision pixel-by-pixel 3D reconstruction method, which uses the grayscale change clues provided by images under different illumination directions to restore the normal direction of the object surface. Photometric stereo has an irreplaceable position in many high-precision 3D reconstruction tasks. For example, it has important application value in archaeological exploration, pipeline inspection, and fine seabed mapping. [0003] However, the existing photometric stereo method based on deep learning has a large error in the high-frequency areas of the object surface, such as wrinkles and...

AI Technical Summary

Problems solved by technology

[0003] However, the existing photometric stereo method based on deep learning has a large error in the high-frequency areas of the object surface, such as wrinkles and edges. The existing methods will generate blurred 3D reconstruction results in these areas. However, these areas are the focus and focus. where accurate reconstruction is required

Images