Creating and Chaining Camera Moves for Quadrotor Videography

ACM Transactions on Graphics (Proceedings of SIGGRAPH 2018)

KE XIE1        HAO YANG1       SHENGQIU HUANG1        DANI LISCHINSKI2         MARC CHRISTIE3        KAI XU1,4        MINGLUN GONG5        DANIEL COHEN-OR1,6        HUI HUANG1*

1Shenzhen University           2The Hebrew University of Jerusalem           3IRISA/INRIA Rennes Bretagne

 4National University of Defense Technology            5Memorial University of Newfoundland               6Tel Aviv University

Fig. 1. Given several landmarks (red areas on the satellite image) and a pair of start and end views (brown camera icons), we generate a suitable camera move (in yellow) for capturing each landmark, and optimally connect them into a continuous and smooth path using transition trajectories (in blue).


Capturing aerial videos with a quadrotor-mounted camera is a challenging creative task, as it requires the simultaneous control of the quadrotor’s motion and the mounted camera’s orientation. Letting the drone follow a pre-planned trajectory is a much more appealing option, and recent research has proposed a number of tools designed to automate the generation of feasible camera motion plans; however, these tools typically require the user to specify and edit the camera path, for example by providing a complete and ordered sequence of key viewpoints.

In this paper, we propose a higher level tool designed to enable even novice users to easily capture compelling aerial videos of large-scale outdoor scenes. Using a coarse 2.5D model of a scene, the user is only expected to specify starting and ending viewpoints and designate a set of landmarks, with or without a particular order. Our system automatically generates a diverse set of candidate local camera moves for observing each landmark, which are collision-free, smooth, and adapted to the shape of the landmark. These moves are guided by a landmark-centric view quality field, which combines visual interest and frame composition. An optimal global camera trajectory is then constructed that chains together a sequence of local camera moves, by choosing one move for each landmark and connecting them with suitable transition trajectories. This task is formulated and solved as an instance of the Set Traveling Salesman Problem.

Fig. 2. A 2.5D model of a large-scale scene with landmarks highlighted in red. Top: several candidate moves generated for each of the four landmarks. Bottom:a continuous drone trajectory generated by chaining together the most suitable local trajectories (in yellow) using smooth transition trajectories (in blue).

Fig. 13. Sunny Beach: this flyby has four landmarks (highlighted in red) of very different sizes. Please see the supplementary video.

Fig. 14. Sea World: this flyby has five landmarks (highlighted in red), where four of them are very close to each other. Please see the supplementary video.



We thank the anonymous reviewers for their valuable comments. This work was supported in part by NSFC (61522213, 61761146002, 6171101466, 61572507), China Postdoc Foundation (2017M622780), 973 Program (2015CB352501), Guangdong Science and Technology Program (2015A030312015), Shenzhen Innovation Program (KQJSCX20170727101233642, JCYJ20151015151249564), ISF-NSFC Joint Research Program (2472/17, 2217/15) and NSERC (293127).


title = {Creating and Chaining Camera Moves for Quadrotor Videography},
author = {Ke Xie and Hao Yang and Shengqiu Huang and Dani Lischinski and Marc Christie and Kai Xu and Minglun Gong and Daniel Cohen-Or and Hui Huang},
journal = {ACM Transactions on Graphics (Proc. SIGGRAPH)},
volume = {37},
number = {4},
pages = {88:1--88:13},  
year = {2018},

Downloads (faster for people in China)

Downloads (faster for people in other places)