Ancient cultural relic three-dimensional reconstruction method based on single view and deep learning

Abstract

The invention relates to the technical field of three-dimensional reconstruction of ancient cultural relic pictures, and discloses an ancient cultural relic three-dimensional reconstruction method based on a single view and deep learning, and the method comprises the following steps: 1, inputting a data set of an ancient cultural relic and a handicraft similar to the ancient cultural relic; 2, according to an encoder-decoder network structure, an encoder with a 3D-ResVGG network and a multi-path channel attention module is used for carrying out deep information mining and feature extraction on the data set, and an ancient cultural relic three-dimensional network model is generated; 3, preprocessing the ancient cultural relic picture needing to be subjected to three-dimensional reconstruction through an intelligent AI tool, and generating a preset type of picture; and step 4, inputting the preset type of picture as a single view into the three-dimensional network model of the ancient cultural relic, and generating a complete three-dimensional model of the ancient cultural relic picture needing three-dimensional reconstruction. The invention provides a new 3D-ResVGG network, and the problem of overfitting of an original VGG architecture is solved through multi-branch model training.

Description

technical field [0001] The invention relates to the technical field of three-dimensional reconstruction of pictures of ancient cultural relics, in particular to a method for three-dimensional reconstruction of ancient cultural relics based on single-view and deep learning. Background technique [0002] Ancient cultural relics are relics and relics with historical, artistic and scientific value left by human beings in social activities. They are precious historical and cultural heritage of human beings. In many cases, people's understanding of ancient cultural relics is only observed through pictures, so the understanding of ancient cultural relics will not be deep enough. Because we live in a three-dimensional world, but what the human eye sees is the two-dimensional projection of the object, and the form of the object obtained through the camera and other equipment is usually a two-dimensional image. Therefore, the three-dimensional reconstruction based on the image is a co...

AI Technical Summary

Problems solved by technology

[0007] Aiming at the current problems of scarcity of views and pictures of ancient cultural relics and scarcity of 3D model data sets, on the basis of existing research and the realization of 3D model data sets of ancient relics, a 3D reconstruction method of ancient cultural relics based on single view and deep learning is proposed. After the 3D network model training of ancient cultural relics is completed, the model can be generated by inputting a single-view image to realize the mapping between the 2D image and the 3D model, so that 3D reconstruction can be performed in the case of a single-view image, and the algorithm does not require image annotation or classification labels for training, overcoming problems such as lack of texture and wide baseline feature matching that could not be solved in the past

Images