Application of a Random Forest Algorithm in Natural Landscape Animation Design

Abstract

Natural landscape simulation is one of the most popular research contents in computer graphics in the field of research simulation system. The natural landscape animation scene can immerse viewers in the scene, and it is widely used in visual simulation systems. Simulating natural scenery on a computer is a powerful method for studying the rules of the scenery's growth process as well as the mystery of life. The simulation of natural scenery is of particular importance and has far-reaching implications. The most important aspect of optimizing natural landscape design is creating a natural landscape animation that users enjoy. This article proposes a natural landscape animation design method with a self-learning function based on this concept. The random forest model (RF) is introduced in this method and applied to the entire animation design process. RF can generate a learning model with user evaluation as the classification result to guide the automatic design of natural landscape animation, resulting in user-satisfying animations. Simultaneously, the RF-based natural landscape animation design can continuously update the learning model based on user needs and is self-learning. The experimental part of this article verifies the effectiveness of the natural landscape animation design proposed in this article by comparing the selection rate of user satisfaction and dissatisfaction scenes, and further demonstrates that the method in this article can improve the natural landscape. The market application value of user satisfaction generated by animation is high.

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

“…Zhao and Zhang [23] have presented application of a RFA in NLAD. Based on such idea, natural landscape animation design technique presented has self-learning capability.…”