吉林大学学报(工学版) ›› 2021, Vol. 51 ›› Issue (6): 2154-2163.doi: 10.13229/j.cnki.jdxbgxb20200671

• 计算机科学与技术 • 上一篇    

基于电影视觉特性的动态多目标实时相机规划

林俊聪(),雷钧,陈萌,郭诗辉(),高星,廖明宏   

  1. 厦门大学 信息学院,福建 厦门 361005
  • 收稿日期:2020-09-03 出版日期:2021-11-01 发布日期:2021-11-15
  • 通讯作者: 郭诗辉 E-mail:jclin@xmu.edu.cn;guoshihui@xmu.edu.cn
  • 作者简介:林俊聪(1981-),男,教授,博士. 研究方向:计算机图形学. E-mail:jclin@xmu.edu.cn
  • 基金资助:
    国家自然科学基金项目(61702433)

Real⁃time camera planning for dynamic multiple targets considering cinematographic visual properties

Jun-cong LIN(),Jun LEI,Meng CHEN,Shi-hui GUO(),Xing GAO,Ming-hong LIAO   

  1. School of Informatics,Xiamen University,Xiamen 361005,China
  • Received:2020-09-03 Online:2021-11-01 Published:2021-11-15
  • Contact: Shi-hui GUO E-mail:jclin@xmu.edu.cn;guoshihui@xmu.edu.cn

摘要:

针对已有相机规划方案应对目标较少或无法做到实时性的问题,从自动实现相机视觉效果优化的角度出发,提出了一种对于动态多目标的实时相机规划方法。首先,通过在多个目标之间构造椭球形的相机空间;之后,获取场景中各目标相对相机的视觉特性,以此来构造非线性规划优化函数。通过模型预测控制的方法对目标在相机屏幕上的表现进行实时优化。通过对算法的运算时间、运行效率以及在场景中的运行表现进行实验,验证了本文方法能够有效地优化场景中各目标的视觉效果,对多目标相机规划问题的研究有一定促进作用。

关键词: 计算机应用, 相机规划, 多目标椭球相机空间, 视觉特性, 非线性优化

Abstract:

As the existing camera planning schemes are unable to deal with multiple targets and achieve real-time effect, in this paper, focusing on automatically optimizing the camera visual effect, a real-time camera planning algorithm is proposed for dynamic multiple targets. The method first adopts the toric camera space among the multiple targets, then takes the visual properties of the targets as the input of the non-linear optimization function. With the model predictive control, we can optimize the targets visual effect in real time. By analyzing the operation time, efficiency and performance of the experiments, it is verified that this method can effectively optimize the visual effect of the targets in the scene and it promotes the research of the multi-target camera planning algorithm.

Key words: computer application, camera planning, toric camera space for multiple targets, visual properties, non-linear optimization

中图分类号: 

  • TP391

图1

相机规划系统流程图"

图2

相机空间图"

图3

相机屏幕投影图"

图4

目标遮挡图"

图5

相机和目标轨迹图"

图6

非线性优化函数运算曲线图"

图7

各视觉特性测试图"

表1

相机空间生成数和预计算时间对比"

目标数量方法生成相机空间数预计算时间/ms
1文献[51411
本文11
2文献[51413
本文13
3文献[514311
本文13
7文献[5142116
本文14
12文献[5146625
本文15

图8

系统实验效果图"

表2

各实验FPS范围值"

实验编号FPS最小值/帧FPS最大值/帧
36.042.9
37.241.3
38.540.1
42.446.8
46.650.2
1 Galvane Q, Lino C, Christie M, et al. Directing cinematographic drones[J]. ACM Transactions on Graphics, 2018, 37(3):1-18.
2 Dong S, Xu K, Zhou Q, et al. Multi-robot collaborative dense scene reconstruction[J]. ACM Transactions on Graphics, 2019, 38(4):1-16.
3 Xu K, Shi Y, Zheng L, et al. 3D attention-driven depth acquisition for object identification[J]. ACM Transactions on Graphics, 2016, 35(6):1-14.
4 Lino C, Christie M. Intuitive and efficient camera control with the toric space[J]. ACM Transactions on Graphics, 2015, 34(4): No.82.
5 Galvane Q, Christie M, Lino C, et al. Camera-on-rails: auto-mated computation of constrained camera paths[C]∥Proceedings of the 8th ACM SIGGRAPH Conference on Motion in Games, Paris, France, 2015: 151-157.
6 Assa J, Caspi Y,Cohen-Or D. Action synopsis: pose selection and illustration[J]. ACM Transactions on Graphics, 2005, 24(3): 667-676.
7 任庆东, 王艺萤, 刘贤梅. 培训系统中虚拟相机运动规划[J]. 科学技术与工程, 2012, 12(8):1936-1940.
Ren Qing-dong, Wang Yi-ying, Liu Xian-mei. Virtual camera motion planning in the simulation training system[J]. Science Technology and Engineering, 2012, 12(8): 1936-1940.
8 Yeh I C, Lin W C, Lee T Y, et al. Social-event-driven camera control for multicharacter animations[J]. IEEE Transactions on Visualization & Computer Graphics, 2012, 18(9): 1496-1510.
9 Li T Y, Xiao X Y. An interactive camera planning system for automatic cinematographer[C]∥Proceedings of 11th International Multimedia Modelling Conference, Melbourne, Australia, 2005: 310-315.
10 Nägeli T, Meier L, Domahidi A, et al. Real-time planning for automated multi-view drone cinematography[J]. ACM Transactions on Graphics, 2017, 36(4): 1-10.
11 Nägeli T, Alonso-Mora J, Domahidi A, et al. Real-time motion planning for aerial videography with dynamic obstacle avoidance and viewpoint optimization[J]. IEEE Robotics and Automation Letters, 2017, 2(3): 1696-1703.
12 丛岩峰, 安向京, 陈虹, 等. 基于滚动优化原理的类车机器人路径跟踪控制[J]. 吉林大学学报:工学版, 2012, 42(1): 182-187.
Cong Yan-feng, An Xiang-jing, Chen Hong, et al. Path following control of car-like robot based on rolling windows[J]. Journal of Jilin University(Engineering and Technology Edition), 2012, 42(1): 182-187.
13 张家旭,王欣志,赵健,等. 汽车高速换道避让路径规划及离散滑模跟踪控制[J]. 吉林大学学报:工学版, 2021, 51(3): 1081-1090.
Zhang Jia-xu, Wang Xin-zhi, Zhao Jian, et al. Path planning and discrete sliding mode tracking control for high⁃speed lane changing collision avoidance of vehicle[J]. Journal of Jilin University(Engineering and Technology Edition), 2021, 51(3): 1081-1090.
14 Lino C, Christie M. Efficient composition for virtual camera control[C]∥Proceedings of the ACM SIGGRAPH/Eurographics Symposium on Computer Animation, Lausanne, Switzerland, 2012: 65-70.
15 Gardner J S, Fowlkes C, Nothelfer C, et al. Exploring aesthetic principles of spatial composition through stock photography[J]. Journal of Vision, 2010, 8(6): 337.
16 Ma Shuang, Fan Yang-yu, Chen Chang-wen. Finding your spot: a photography suggestion system for placing human in the scene[C]∥Proceedings of IEEE International Conference on Image Processing, Paris, France, 2014: 556-560.
[1] 曹洁,屈雪,李晓旭. 基于滑动特征向量的小样本图像分类方法[J]. 吉林大学学报(工学版), 2021, 51(5): 1785-1791.
[2] 孙小雪,钟辉,陈海鹏. 基于决策树分类技术的学生考试成绩统计分析系统[J]. 吉林大学学报(工学版), 2021, 51(5): 1866-1872.
[3] 张萌谡,刘春天,李希今,黄永平. 基于K⁃means聚类算法的绩效考核模糊综合评价系统设计[J]. 吉林大学学报(工学版), 2021, 51(5): 1851-1856.
[4] 姚引娣,贺军瑾,李杨莉,谢荡远,李英. 自构建改进型鲸鱼优化BP神经网络的ET0模拟计算[J]. 吉林大学学报(工学版), 2021, 51(5): 1798-1807.
[5] 赵宏伟,张子健,李蛟,张媛,胡黄水,臧雪柏. 基于查询树的双向分段防碰撞算法[J]. 吉林大学学报(工学版), 2021, 51(5): 1830-1837.
[6] 王春波,底晓强. 基于标签分类的云数据完整性验证审计方案[J]. 吉林大学学报(工学版), 2021, 51(4): 1364-1369.
[7] 欧阳丹彤,刘扬,刘杰. 故障响应指导下基于测试集的故障诊断方法[J]. 吉林大学学报(工学版), 2021, 51(3): 1017-1025.
[8] 钱榕,张茹,张克君,金鑫,葛诗靓,江晟. 融合全局和局部特征的胶囊图神经网络[J]. 吉林大学学报(工学版), 2021, 51(3): 1048-1054.
[9] 朱小龙,谢忠. 基于机器学习的地理空间数据抽取算法[J]. 吉林大学学报(工学版), 2021, 51(3): 1011-1016.
[10] 孙宝凤,任欣欣,郑再思,李国一. 考虑工人负荷的多目标流水车间优化调度[J]. 吉林大学学报(工学版), 2021, 51(3): 900-909.
[11] 刘元宁,吴迪,朱晓冬,张齐贤,李双双,郭书君,王超. 基于YOLOv3改进的用户界面组件检测算法[J]. 吉林大学学报(工学版), 2021, 51(3): 1026-1033.
[12] 陈广秋,陈昱存,李佳悦,刘广文. 基于DNST和卷积稀疏表示的红外与可见光图像融合[J]. 吉林大学学报(工学版), 2021, 51(3): 996-1010.
[13] 王淑敏,陈伟. 基于连续密度隐马尔可夫模型的矿下异常行为识别算法[J]. 吉林大学学报(工学版), 2021, 51(3): 1067-1072.
[14] 沈淑涛,尼玛扎西. 基于区块链技术的双混沌可识篡改图像加密方法[J]. 吉林大学学报(工学版), 2021, 51(3): 1055-1059.
[15] 周炳海,何朝旭. 基于静态半成套策略的多目标准时化物料配送调度[J]. 吉林大学学报(工学版), 2021, 51(3): 910-916.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!