OVERVIEW
STATEMENT OF THE PROBLEM
TARGET AUDIENCE
RESULT
The most critical purpose of art graphic design is to solve thepositioning solution and emergency plan of the product inthe complete solution. During the establishment of the finalsolution of the product, different design processes andemergency measures need to be presented in visual com-munication in real time [1, 2]. Based on the artistic designneeds of the client’s feedback, the appearance and approachare reconstructed in terms of appearance, methodologicalimprovement, and style review on different levels such asscreen, space, structure, and logic. In the field of visualdesign, textual language is active in another way in thecoordination of solutions.
Artistic graphic design is the aesthetic result of the designer’s fusion of various elements, with a high degree of independence. Considering the lack of significant visual design scope and aesthetic indicators of graphic design, our research aims to build an upgraded network model that can categorize different types of artistic graphics with labels and realize the free combination of graphic solutions. We realize the scheme reorganization of artistic graphic design from the perspective of computer vision and propose the artistic graphic design method based on memory neural network.
Considering the high professional require-ments of artistic graphic design for manual labor, manyresearchers started to study automatic combination systemsfor artistic graphic design [4–7]. is research requires theintegration of an artistic graphics database, computer visionunit, deep learning algorithm unit, data preprocessing unit,etc. Many researchers have already started their workaccordingly.
In the last part of the network structure, wepropose the LSTM structure based on the attentionmechanism to match with the self-attention features of thegraphic region segmentation module and pass the matchedattention feature vector to the LSTM network to extract thelabeled text feature information of the graphics. To test theeffectiveness of our method, we build a database of artisticgraphics and set up an adaptive training process. We also compare deep learning methods of the same type, and theexperimental results demonstrate that our method outper-forms other deep methods in artistic graphic design in termsof scheme reorganization accuracy and quantitative evalu-ation of artistic models
Visual Memory Neural Network for Artistic Graphic Design
Author: Yuchuan Zheng
Date: 2022