Font Size: a A A

Research On Automatic Generation Method Of Personalized Shadow Pupperty Modeling Based On Generative Adversarial Network

Posted on:2021-04-11Degree:MasterType:Thesis
Country:ChinaCandidate:Z L GongFull Text:PDF
GTID:2428330614450250Subject:Design
Abstract/Summary:PDF Full Text Request
Shadow puppetry is one of the numerous intangible cultural heritages in China,which has a high artistic appreciation value.However,with the continuous development of The Times,people's entertainment modes have undergone great changes at present.Due to the slow development pace and insufficient novelty,shadow puppetry gradually falls into a situation of extinction.To shadow play as the research object,this paper analyzes the importance of protect the non-material cultural heritage,and the inheritance and development of traditional folk art meaning,finally proposes the personalized shadow play modelling based on emergent against network automatic generation algorithm,realized under the condition of preserving face feature automatic generation shadow play.In this paper,the related research of style transformation at home and abroad is discussed,and the generation of cyclic consistency countermeasure network is introduced in detail,and its advantages and disadvantages are analyzed.Then,based on the analysis results,an algorithm for the automatic generation of personalized shadow puppets is proposed.The method consists of two parts.In the first part,a forward face to face algorithm is proposed,which is mainly used to change the input positive face image into a side face image.The second part is the silhouette stylization of the side face.This part mainly relies on the cyclic consistency countermeasure network,and realizes the shadow shadow style transformation of the side face on the basis of it.This paper compares and analyzes the experimental results of the personalized shadow figure automatic generation method based on the generative confrontation network.At the end of the thesis,a personalized shadow puppetry generation system is designed to process the face image,so that users can get the shadow puppetry image conveniently.This project proposes an algorithm that can retain facial features in shadow puppetry images,and it is expected to attract young people's attention in this way and protect the shadow puppetry art when it is developing.In the end,this paper summarizes the deficiencies and provides guidance for future research.
Keywords/Search Tags:Shadow play, Style change, GAN, CycleGAN
PDF Full Text Request
Related items