A multi-category multi-view data generation method based on deep convolutional generative adversarial network

Abstract

The present invention provides a multi-category multi-view data generation method for generating a confrontation network based on the depth convolution. The method comprises the following steps: 1. carrying out picture center cutting; 2. overlaying multiple views in the channel dimension; 3. extracting multi-view category tags; 4. co-training the DC-GAN network by using the overlaid multiple views, category tags, and high-dimensional random noise; 5.introducing the high-dimensional random noise and custom tags into the trained network, and generating multi-view overlaid data; and 6. cutting and filling the background to obtaining the multiple views satisfying the original size. According to the method provided by the present invention, the method for generating the confrontation network bymulti-view overlaying and tag training realizes the function that the multi-category multi-view can be generated through a model only by modifying the input, and the generated data can be used as extension of trained data to increase the diversity of trained data.

Description

technical field [0001] The present invention relates to the fields of deep learning and image processing, and related data (especially image data) generation technology, especially for single-object picture data of multiple categories and multiple perspectives, such as multiple views of different types of pearls in the pearl industry. Background technique [0002] In recent years, with the continuous development of deep learning technology, great breakthroughs have been made in a series of problems such as classification and object detection, and multi-layer neural network structures emerge in endlessly. However, the more complex the neural network, the higher the demand for the quantity and diversity of training data, and the final performance of the neural network is positively correlated with the richness of the training data in a wide range. [0003] In order to increase the richness of training data, the safest and most reliable method is to manually collect and label o...

AI Technical Summary

Problems solved by technology

[0004] At present, in the generation of image data, more single-image generation is based on large-scale data sets. In some practical object classification problems, such as pearl classification, using multi-view data can achieve better results, but there are One of the problems is that the amount of data is not large and the acquisition cost is high. Therefore, it is an urgent need to generate multi-category and multi-view data

Images