FLASH: Fast Learning Based Automated Scene Hyper-Lapse
Abstract
The increasing availability of visual data from mobile devices and digital platforms has created a pressing need for automated tools capable of transforming raw imagery into meaningful visual content. Hyper-lapse videos offer a compact, visually compelling representation of real-world scenes; however, traditional approaches rely on manual editing or tightly controlled capture setups. This work presents FLASH (Fast Learning-Based Automated Scene Hyper-Lapse), an AI-driven pipeline for fully automated hyper-lapse generation from unconstrained image collections. The system integrates 3D scene reconstruction via Structure-from-Motion (COLMAP) and Neural Radiance Fields (NeRF), with quaternion spline interpolation for smooth, gimbal-lock-free camera trajectory generation. A BayesRays uncertainty-refinement stage optimizes the field of view to suppress floater artifacts. Evaluated on GPS-sampled Google Street View sequences, the approach achieves a 5.25× reduction in inter-frame angular jitter and produces artifact-free stabilized hyper-lapse videos without human intervention, demonstrating suitability for digital content creation, tourism visualization, and automated media production.
References
[1] J. Kopf, M. F. Cohen, and R. Szeliski, “First-person hyper-lapse videos,” ACM Trans. Graph. (SIGGRAPH), vol. 33, no. 4, pp. 78:1–78:10, 2014. https://doi.org/10.1145/2601097.2601195
[2] B. Mildenhall, P. P. Srinivasan, M. Tancik, J. T. Barron, R. Ramamoorthi, and R. Ng, “NeRF: Representing scenes as neural radiance fields for view synthesis,” in Proc. ECCV, 2020. https://doi.org/10.1007/978-3-030-58452-8_24
[3] T. Müller, A. Evans, C. Schied, and A. Keller, “Instant neural graphics primitives with a multiresolution hash encoding,” ACM Trans. Graph. (SIGGRAPH), vol. 41, no. 4, 2022. https://doi.org/10.1145/3528223.3530127
[4] M. Tancik et al., “Nerfstudio: A modular framework for neural radiance field development,” in Proc. ACM SIGGRAPH, 2023. https://doi.org/10.1145/3588432.3591516
[5] J. L. Schönberger and J.-M. Frahm, “Structure-from-motion revisited,” in Proc. CVPR, 2016. https://doi.org/10.1109/CVPR.2016.445
[6] L. Goli, C. Reading, S. Sellán, A. Jacobson, and A. Tagliasacchi, “Bayes’ rays: Uncertainty quantification for neural radiance fields,” in Proc. CVPR, 2024. https://doi.org/10.1109/CVPR52733.2024.00361
[7] K. Shoemake, “Animating rotation with quaternion curves,” ACM SIGGRAPH Comput. Graph., vol. 19, no. 3, pp. 245–254, 1985. https://doi.org/10.1145/325165.325242
[8] Google LLC, “Street View Static API,” Google Developers, 2024. https://developers.google.com/maps/documentation/streetview/overview
Additional Files
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 International Journal of Artificial Intelligence and Modern Engineering
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
