Mirror object three-dimensional reconstruction method, device, computer equipment and storage medium

Abstract

The present application relates to a method, device, computer equipment, storage medium and computer program product for three-dimensional reconstruction of mirror objects, which are applied in the technical field of three-dimensional reconstruction and used to improve the effect of three-dimensional reconstruction of mirror objects. The method includes: inputting the photographed image of the mirror object into the rendering information network model, decoding the pixels in the photographed image through the rendering information network model, and obtaining the reconstruction information and the second Rendering information; input the reconstruction information and the preset viewing angle of the specular object into the implicit scene model for rendering processing, and obtain the second rendering information and the preset viewing angle for the specular object output by the implicit scene model A perspective image is set; according to the captured image, the preset perspective image, the first rendering information, and the second rendering information, perform three-dimensional reconstruction on the specular object to obtain a three-dimensional reconstruction model of the specular object.

Description

technical field [0001] The present application relates to the technical field of three-dimensional reconstruction, and in particular, to a method, apparatus, computer equipment, storage medium and computer program product for three-dimensional reconstruction of a specular object. Background technique [0002] 3D reconstruction refers to extracting spatial information from 2D images or videos, and using spatial information to restore models of 3D objects and scenes. Due to the high precision and complexity of traditional 3D reconstruction techniques, the development of 3D reconstruction techniques has been relatively slow in the past few decades. With the continuous development of deep learning, the field of deep learning technology has also begun to be applied in the field of 3D reconstruction. [0003] Existing 3D reconstruction methods based on deep learning can only be applied to 3D reconstruction of objects with common materials, and lack feature processing for objects ...

AI Technical Summary

Problems solved by technology

For example, the surface of a specular object is highly reflective, and a large amount of reflected light on the surface of a specular object makes it difficult to extract effective features by traditional deep learning methods, which makes the effect of 3D reconstruction of specular objects poor when using traditional methods

Images