Research And Implementation Of Interactive Sketch Painting Teaching System

From ancient times to now,sketches play an important role in human life,so the painting of sketch painting has become one of the contents of children’s primary art education.The traditional offline painting education is constrained by time,place,cost,etc.,resulting in slow progress of students’learning;traditional online computer teaching software is mostly embodied in video form,and teaching interaction is poor.Inspired by the application of existing deep learning techniques in the field of education,this paper combines deep neural network algorithms with traditional teaching systems to solve the shortcomings of current simple painting education.This system aims to use the sequence of the simple drawing process as the teaching content,and attempts to intelligently guide the young students to complete the process of learning painting by the method of human-computer interaction teaching.In a word,the resear-ch content of this paper is divided into four parts:(1)In order to realize the function of teaching and learning in the teaching system,this paper proposes a sequence-generating algorithm for various types of simple stroke painting process based on attention mechanism.The complete stick figure is used as the prior knowledge of generating the sequence of the simple drawing process,and the generation is accurate.The stroke sequence of the drawing process.The Turing test results show that the proposed algorithm has achieved good results.(2)This paper proposes a target recognition algorithm for recognizing the image sequence of the simple stroke painting process.This algorithm uses the current reliable resnet-50 as the feature extraction network,and adds the dropout mechanism to make the recognition part to the complete sequence of the simple stroke painting process.The accuracy reaches 87.49%.(3)In order to realize the teaching function of the out-of-town sketching in real teaching,the conversion function between the stick figure and the real scene picture is added to the intelligent system,so that many young students can understand the real object corresponding to the stick figure they draw.(4)Design and implementation of interactive sketch painting teaching system based on deep neural network.In this paper,the deep neural network model is combined with the traditional teaching system,and the problem of large amount of data in the network model is considered.The teaching system based on deep neural network is designed and implemented.Totally by improving related system algorithm,the target of the thesis is to design and implement an intelligent interactive stick figure teaching system.