数字媒体动画场景细节自动控制方法研究

Research on Automatic control method for scene details of digital media animation

针对当前采用静态平滑处理技术处理动画场景细节,导致出现动画场景细节清晰度低的问题,提出了数字媒体动画场景细节自动控制方法。采用多目标跟踪技术,通过Kalman滤波器的预测和更新两个阶段,跟踪动画场景每一帧的细节信息;将动画场景平滑分为静态场景细节平滑与动态场景细节平滑两部分,平滑处理动画场景细节信息;依据动画场景细节信息,定义三维R树索引技术3个参数,设计动画场景检索算法,检索场景细节自动控制目标,实现数字媒体动画场景细节自动控制。选择一张数字媒体动画中的场景图片,确定细节自动控制清晰度评价计算公式,并设计对比实验,结果表明:本研究方法控制动画场景细节清晰度相较基于改进RBF的控制方法和基于强化学习的控制方法分别高591和355。

An automatic control method for digital media animation scene details is proposed to address low image quality in animation scene details caused by the use of static smoothing techniques.The method employs multi-object tracking technology to track the detail information in each frame of animation scenes through the prediction step and the update step of the Kalman filter.Animation scene smoothing is divided into two parts:static scene detail smoothing and dynamic scene detail smoothing,which are used to process the detail information of animation scenes.Based on the detail information of animation scenes,three parameters of the 3D R-tree index structure are defined,and an animation scene retrieval algorithm is designed to retrieve the automatic control targets of scene details,thus achieving automatic control-of digital media animation scene details.A scene image from a digital media animation is selected,a quality evaluation formula for automatic detail control is determined,and comparative experiments are designed.The results show that the quality of the animation scene details controlled by the proposed method is improved by 591 and 355 respectively compared to the methods based on improved RBF and reinforcement learning.