Miao Wang, Jun-Bang Liang, Song-Hai Zhang, Shao-Ping Lu, Ariel Shamir, Shi-Min Hu
Hyper-lapse video with high speed-up rate is an efficient way to overview long videos, such as a human activity in first-person view. Existing hyper-lapse video creation methods produce a fast-forward video effect using only one video source. In this paper, we present a novel hyper-lapse video creation approach based on multiple spatially-overlapping videos. We assume the videos share a common view or location, and find transition points where jumps from one video to another may occur. We represent the collection of videos using a hyper-lapse transition graph; the edges between nodes represent possible hyper-lapse frame transitions. To create a hyper-lapse video, a shortest path search is performed on this digraph to optimize frame sampling and assembly simultaneously. Finally, we render the hyper-lapse results using video stabilization and appearance smoothing techniques on the selected frames. Our technique can synthesize novel virtual hyper-lapse routes, which may not exist originally. We show various application results on both indoor and outdoor video collections with static scenes, moving objects, and crowds.
Wang, M, Liang, J-B, Zhang, S-H, Lu, S-P, Shamir, A & Hu, S-M 2018, 'Hyper-Lapse From Multiple Spatially-Overlapping Videos', IEEE Transactions on Image Processing, vol. 27, no. 4, 8025571, pp. 1735-1747. https://doi.org/10.1109/TIP.2017.2749143
Wang, M., Liang, J.-B., Zhang, S.-H., Lu, S.-P., Shamir, A., & Hu, S.-M. (2018). Hyper-Lapse From Multiple Spatially-Overlapping Videos. IEEE Transactions on Image Processing, 27(4), 1735-1747. Article 8025571. https://doi.org/10.1109/TIP.2017.2749143
@article{6c10d1ecca3c4b5ca74ee6bcbffabed0,
title = "Hyper-Lapse From Multiple Spatially-Overlapping Videos",
abstract = "Hyper-lapse video with high speed-up rate is an efficient way to overview long videos, such as a human activity in first-person view. Existing hyper-lapse video creation methods produce a fast-forward video effect using only one video source. In this paper, we present a novel hyper-lapse video creation approach based on multiple spatially-overlapping videos. We assume the videos share a common view or location, and find transition points where jumps from one video to another may occur. We represent the collection of videos using a hyper-lapse transition graph; the edges between nodes represent possible hyper-lapse frame transitions. To create a hyper-lapse video, a shortest path search is performed on this digraph to optimize frame sampling and assembly simultaneously. Finally, we render the hyper-lapse results using video stabilization and appearance smoothing techniques on the selected frames. Our technique can synthesize novel virtual hyper-lapse routes, which may not exist originally. We show various application results on both indoor and outdoor video collections with static scenes, moving objects, and crowds.",
keywords = "Hyper-lapse video, time-lapse, video editing, video synthesis",
author = "Miao Wang and Jun-Bang Liang and Song-Hai Zhang and Shao-Ping Lu and Ariel Shamir and Shi-Min Hu",
note = "IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 27, NO. 4, APRIL 2018, pp 1735-1747",
year = "2018",
month = apr,
day = "1",
doi = "10.1109/TIP.2017.2749143",
language = "English",
volume = "27",
pages = "1735--1747",
journal = "IEEE Transactions on Image Processing",
issn = "1057-7149",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "4",
}