Efficiently Reconstructing Dynamic Scenes One D4RT at a Time
Chuhan Zhang, Guillaume Le Moing, Skanda Koppula, Ignacio Rocco, Liliane Momeni, Junyu Xie, Shuyang Sun, Rahul Sukthankar
Abstract
Understanding and reconstructing the complex geometry and motion of dynamic 4D scenes from video remains a formidable challenge in computer vision. This paper introduces D4RT, a simple yet powerful feedforward network designed to efficiently solve this task. D4RT utilizes a unified transformer architecture to jointly infer depth, spatio-temporal correspondence, and full camera parameters from a single video. Its core innovation is a novel mechanism that sidesteps the heavy computation of dense, per-frame decoding and the complexity of managing multiple, task-specific decoders. Our unified decoding interface allows the model to independently and efficiently probe the 3D position of any point in space and time. The result is a lightweight and highly scalable method that enables remarkably efficient training and inference. We demonstrate that our approach sets a new state-of-the-art, outperforming previous methods across a wide spectrum of 4D reconstruction tasks.
BibTeX
@inproceedings{cvpr2026_efficientlyrecon,
title = {Efficiently Reconstructing Dynamic Scenes One D4RT at a Time},
author = {Chuhan Zhang and Guillaume Le Moing and Skanda Koppula and Ignacio Rocco and Liliane Momeni and Junyu Xie and Shuyang Sun and Rahul Sukthankar and Joëlle K. Barral and Raia Hadsell and Zoubin Ghahramani and Andrew Zisserman and Junlin Zhang and Mehdi S. M. Sajjadi},
booktitle = {CVPR 2026},
year = {2026}
}