Artistic Creation and Animation Scene Designing using Hierarchical Spatio-Temporal Graph Convolutional Neural Network Optimized with Gazelle Optimization Algorithm

Abstract

Creating moving visuals, usually in the form of 2D or 3D animations, is the focus of the graphic design field known as animation design. A sequence of consecutive images, or frames, is played quickly one after the other in animation design to provide the impression of motion. In this Manuscript, Artistic Creation and Animation Scene Designing using Hierarchical Spatio-temporal graph convolutional neural network Optimized with Gazelle Optimization (AC-ASD-HSTCNN-GO)is proposed. Initially, data is taken from LiDAR dataset then the data are fed to preprocessing segment. For pre-processing Orthogonal Master Slave Adaptive Notch Filter (OMSANF) is used to remove duplicate data and replacing missing data from collected data’s. Then the feature are extracted by using General Synchroextracting Chirplet Transform (GSCT) technique, which extracts features such as objects, faces and scenes. Finally HSTGCNN for creating 3D animation image design .In general the HSTGCNN does not express some adaption of optimization strategies for determining optimum parameters to assure the product recommendation. Hence, the GOA is proposed to enhance HSTGCNN.The proposed AC-ASD-HSTCNN-GO method is activated in python and the performance of proposed AC-ASD-HSTCNN-GO is estimated under performance metrics like accuracy, precision, F1-score, recall, specificity, computational time, error rate, RoC. Finally, performance of AC-ASD-HSTCNN-GO method provides 15.27%, 17.14%, 18.58% greater accuracy, 16.11%, 19.65%, 20.53% greater precision and 19.31%, 21.29%, 22.15% greater F1-Score while compared with existing techniques like Graphic Design of 3D Animation Scenes Depend on Deep Learning with Information Security Technology (GD-3DASDL-IST), GRADE: Generating Realistic Animated Dynamic Environments for Robotics Research (GRADE-RR)andApplication of Random Forest Algorithm in Natural Landscape Animation Design (ALN-RFA-NLAD) respectively.