Figure image generation method and device

• Main Authors: LIU ZHICHAO, GAO CHEN, YU LIANGHUAN, WU ZHUOZHUO
• Format: Patent
• Language: Chinese; English
• Published: 12.06.2020
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

The invention provides a figure image generation method and device, and the method comprises the steps: firstly obtaining a description text; wherein the description text comprises description information of a target person image; secondly, generating a semantic feature vector corresponding to the description text; performing operation on the semantic feature vector to obtain a constant vector ina standard format; wherein the operation process of the semantic feature vector is performed by adopting a full connection layer; and finally, inputting the constant vector in the standard format intoa generative adversarial network model, and performing operation by the generative adversarial network model to obtain and output the target person image. Due to the fact that the generative adversarial network model and the full connection layer are optimal models obtained through multiple times of training, in the implementation process of the method, the corresponding figure image can be accurately generated through th