Online3R
Online Learning for Consistent Sequential Reconstruction Based on Geometry Foundation Model

1School of Intelligence Science and Technology, Peking University 2State Key Laboratory of General Artificial Intelligence 3T-Stone Robotics Institute, The Chinese University of Hong Kong 4NVIDIA 5College of Computer and Information Science, Southwest University 6School of Artificial Intelligence and Computer Science, Anqing Normal University
CVPR 2026
Online3R Teaser/Pipeline

The overall pipeline of Online3R. The core of our Online3R lies in constructing self-supervised methods and updating prompts online based on supervision signals, enabling the model to adapt to the current scene and ensuring consistent reconstruction results. We leverages a local consistency loss derived from temporally fused geometry to improve immediate accuracy, and a global consistency loss that enforces geometric invariance across distant views to mitigate long-term drift, guaranteeing both consistency and efficiency.

Abstract

We present Online3R, a new sequential reconstruction framework that is capable of adapting to new scenes through online learning, effectively resolving inconsistency issues. Specifically, we introduce a set of learnable lightweight visual prompts into a pretrained, frozen geometry foundation model to capture the knowledge of new environments while preserving the fundamental capability of the foundation model for geometry prediction. To solve the problems of missing groundtruth and the requirement of high efficiency when updating these visual prompts at test time, we introduce a local-global self-supervised learning strategy by enforcing the local and global consistency constraints on predictions. The local consistency constraints are conducted on intermediate and previously local fused results, enabling the model to be trained with high-quality pseudo groundtruth signals; the global consistency constraints are operated on sparse keyframes spanning long distances rather than per frame, allowing the model to learn from a consistent prediction over a long trajectory in an efficient way. Our experiments demonstrate that Online3R outperforms previous state-of-the-art methods on various benchmarks.

BibTeX

@article{YourPaperKey2024,
  title={online3r},
  author={Zhou, Shunkai and Yan, Zike and Xue, Fei and Wu, Dong and Deng, Yuchen and Zha, Hongbin},
  journal={CVPR},
  year={2026}
}