4DGT: Learning a 4D Gaussian Transformer Using Real-World Monocular Videos
This addresses the problem of efficient and accurate 4D scene reconstruction for applications in computer vision and robotics, representing a novel method for a known bottleneck.
The paper tackles dynamic scene reconstruction from monocular videos by proposing 4DGT, a 4D Gaussian-based Transformer model that unifies static and dynamic components, reducing reconstruction time from hours to seconds and achieving on-par accuracy with optimization-based methods on cross-domain videos.
We propose 4DGT, a 4D Gaussian-based Transformer model for dynamic scene reconstruction, trained entirely on real-world monocular posed videos. Using 4D Gaussian as an inductive bias, 4DGT unifies static and dynamic components, enabling the modeling of complex, time-varying environments with varying object lifespans. We proposed a novel density control strategy in training, which enables our 4DGT to handle longer space-time input and remain efficient rendering at runtime. Our model processes 64 consecutive posed frames in a rolling-window fashion, predicting consistent 4D Gaussians in the scene. Unlike optimization-based methods, 4DGT performs purely feed-forward inference, reducing reconstruction time from hours to seconds and scaling effectively to long video sequences. Trained only on large-scale monocular posed video datasets, 4DGT can outperform prior Gaussian-based networks significantly in real-world videos and achieve on-par accuracy with optimization-based methods on cross-domain videos. Project page: https://4dgt.github.io