UAV4D: Dynamic Neural Rendering of Human-Centric UAV Imagery Using Gaussian Splatting
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (5) · 3372-3380
DOI: 10.1609/aaai.v40i5.37333 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Jaehoon Choi und 6 weiteren Autor:innen.
Abstract (Originalsprache)
Despite significant advancements in dynamic neural rendering, existing methods fail to address the unique challenges posed by UAV-captured scenarios, particularly those involving monocular camera setups, top-down perspective, and multiple small, moving humans, which are not adequately represented in existing datasets. In this work, we introduce UAV4D, a framework for enabling photorealistic rendering for dynamic real-world scenes captured by UAVs. Specifically, we address the challenge of reconstructing dynamic scenes with multiple moving pedestrians from monocular video data without the need for additional sensors. We use a combination of a 3D foundation model and a human mesh reconstruction model to reconstruct both the scene background and humans. We propose a novel approach to resolve the scene scale ambiguity and place both humans and the scene in world coordinates by identifying human-scene contact points. Additionally, we exploit the SMPL model and background mesh to initialize Gaussian splats, enabling holistic scene rendering. We evaluated our method on three complex UAV-captured datasets: VisDrone, Manipal-UAV, and Okutama-Action, each with distinct characteristics and 10-50 humans. Our results demonstrate the benefits of our approach over existing methods in novel view synthesis, achieving a 1.5 dB PSNR improvement and superior visual sharpness.
Zitieren
APA
Choi, J., Jung, D., Maxey, C., Eum, S., Lee, Y., Manocha, D., & Kwon, H. (2026). UAV4D: Dynamic Neural Rendering of Human-Centric UAV Imagery Using Gaussian Splatting. Proceedings of the AAAI Conference on Artificial Intelligence, 40(5), 3372-3380. https://doi.org/10.1609/aaai.v40i5.37333
Vancouver
Choi J, Jung D, Maxey C, Eum S, Lee Y, Manocha D, et al. UAV4D: Dynamic Neural Rendering of Human-Centric UAV Imagery Using Gaussian Splatting. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(5):3372-3380. doi:10.1609/aaai.v40i5.37333
BibTeX
@article{Choi2026,
author = {Choi, Jaehoon and Jung, Dongki and Maxey, Chris and Eum, Sungmin and Lee, Yonghan and Manocha, Dinesh and Kwon, Heesung},
title = {UAV4D: Dynamic Neural Rendering of Human-Centric UAV Imagery Using Gaussian Splatting},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {5},
pages = {3372--3380},
doi = {10.1609/aaai.v40i5.37333}
}
EUVIMED