Nanchang Hangkong University

Congxuan Zhang

Zhen Chen

Ming Li

Hao Chen

Leqi Jiang

Zige Wang

Cheng Feng

Chengzhong Wu

Weiming Hu

Yaonan Wang

Shijie Zhang

Liyue Ge

A method for generating an industrial defect image, a device, a medium, and a product are provided, relating to the field of image processing. The method includes: acquiring an industrial defect-free image; and generating an industrial defect image based on the industrial defect-free image by using a defect image generation model. The defect image generation model is constructed based on a cyclic Generative Adversarial Network (GAN); the defect image generation model includes two generators and two discriminators, with each generator including an encoder, a transformer, and a decoder connected in sequence; the encoder uses convolution operations to extract features from an input image, the transformer uses a self-attention-based residual network to transform an image output by the encoder, and the decoder uses a skip connection mechanism and deconvolution operations to decode an image output by the transformer to obtain a generated image.

Method for generating industrial defect image, device, medium, and product

A method and system for optimizing optical flow for images based on a residual field and a displacement field are provided. The method includes: obtaining reference images; estimating an initial optical flow field from the reference images by using an optical flow estimation method; obtaining an optical flow optimization model, where the optical flow optimization model includes an image encoder, an optical flow encoder, a first decoder, and a sub-decoder; inputting any image of the reference images and the initial optical flow field into the optical flow optimization model to output the residual field and the displacement field; superimposing the initial optical flow field and the residual field to obtain a preliminarily optimized optical flow field; and resampling the preliminarily optimized optical flow field by using the displacement field to obtain an optimized optical flow field.

Method and system for optimizing optical flow for images based on residual field and displacement field

Growing community of inventors

Nanchang, China

Congxuan Zhang