刘春雷

演播室多机位节目制作长期依赖人工导播完成镜头选取与切换，人力成本高且质量难以量化。为此提出一种融合身份感知与时空约束的跨镜头人物跟踪及导播决策方法，构建由感知、单镜头跟踪、身份确认、跨镜头关联与导播决策组成的五层框架：以限制对比度自适应直方图均衡抑制强补光造成的照度失衡；以人脸质量筛选、局部二值模式粗筛与轻量卷积嵌入精判构成的级联结构建立稳定身份；将地面单应投影一致性、运动时空门控与再识别外观相似度整合为自适应加权代价，经两级匈牙利分配实现弱重叠及无重叠视域下的全局身份传递；并以占屏比、构图偏移、人脸朝向、清晰度与运动平稳度五要素构图评分结合最小驻留时长与迟滞阈值的切换规则输出播出画面。经实践检验，该方法具备实时辅助导播能力。

多摄像头多目标跟踪；跨镜头身份关联；时空约束；自动导播；镜头切换决策

[1] Zhang Y F,Sun P Z,Jiang Y,et al.ByteTrack:multi-object tracking by associating every detection box[C]//Proceedings of the European Conference on Computer Vision.Cham:Springer,2022:1-21. [2] Luiten J,Ošep A,Dendorfer P,et al.HOTA:a higher order metric for evaluating multi-object tracking[J].International Journal ofComputer Vision,2021,129(2):548-578. [3] Tang Z,Naphade M,Liu M Y,et al.CityFlow:a city-scale benchmark for multi-target multi-camera vehicle tracking and re-identification[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway:IEEE,2019:8797-8806. [4] Tanabe Y,Nakanishi T.Learning-based scene selection for autonomous multi-camera video editing[C]//Proceedings of the 19th IIAI International Congress on Advanced Applied Informatics.Piscataway:IEEE,2025:155-160.

点击下载PDF