Feed-Forward Bullet-Time Reconstruction of Dynamic Scenes from Monocular Videos
Hanxue Liang, Jiawei Ren, Ashkan Mirzaei, Antonio Torralba, Ziwei Liu, Igor Gilitschenski, Sanja Fidler, Cengiz Oztireli
Abstract
Recent advancements in static feed-forward scene reconstruction have demonstrated significant progress in high-quality novel view synthesis. However, these models often struggle with generalizability across diverse environments and fail to effectively handle dynamic content. We present BTimer (short for Bullet Timer), the first motion-aware feed-forward model for real-time reconstruction and novel view synthesis of dynamic scenes. Our approach reconstructs the full scene in a 3D Gaussian Splatting representation at a given target (‘bullet’) timestamp by aggregating information from all the context frames. Such a formulation allows BTimer to gain scalability and generalization by leveraging both static and dynamic scene datasets. Given a casual monocular dynamic video, BTimer reconstructs a bullet-time scene within 150ms while reaching state-of-the-art performance on both static and dynamic scene datasets, even compared with optimization-based approaches.
BibTeX
@inproceedings{
liang2025feedforward,
title={Feed-Forward Bullet-Time Reconstruction of Dynamic Scenes from Monocular Videos},
author={Hanxue Liang and Jiawei Ren and Ashkan Mirzaei and Antonio Torralba and Ziwei Liu and Igor Gilitschenski and Sanja Fidler and Cengiz Oztireli and Huan Ling and Zan Gojcic and Jiahui Huang},
booktitle={The Thirty-ninth Annual Conference on Neural Information Processing Systems},
year={2025},
url={https://openreview.net/forum?id=oGc1qHAUBJ}
}