LONG3R: Long Sequence Streaming 3D Reconstruction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Zhuoguang, Qin, Minghui, Yuan, Tianyuan, Liu, Zhe, Zhao, Hang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911074604810240
author Chen, Zhuoguang
Qin, Minghui
Yuan, Tianyuan
Liu, Zhe
Zhao, Hang
author_facet Chen, Zhuoguang
Qin, Minghui
Yuan, Tianyuan
Liu, Zhe
Zhao, Hang
contents Recent advancements in multi-view scene reconstruction have been significant, yet existing methods face limitations when processing streams of input images. These methods either rely on time-consuming offline optimization or are restricted to shorter sequences, hindering their applicability in real-time scenarios. In this work, we propose LONG3R (LOng sequence streaming 3D Reconstruction), a novel model designed for streaming multi-view 3D scene reconstruction over longer sequences. Our model achieves real-time processing by operating recurrently, maintaining and updating memory with each new observation. We first employ a memory gating mechanism to filter relevant memory, which, together with a new observation, is fed into a dual-source refined decoder for coarse-to-fine interaction. To effectively capture long-sequence memory, we propose a 3D spatio-temporal memory that dynamically prunes redundant spatial information while adaptively adjusting resolution along the scene. To enhance our model's performance on long sequences while maintaining training efficiency, we employ a two-stage curriculum training strategy, each stage targeting specific capabilities. Experiments demonstrate that LONG3R outperforms state-of-the-art streaming methods, particularly for longer sequences, while maintaining real-time inference speed. Project page: https://zgchen33.github.io/LONG3R/.
format Preprint
id arxiv_https___arxiv_org_abs_2507_18255
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LONG3R: Long Sequence Streaming 3D Reconstruction
Chen, Zhuoguang
Qin, Minghui
Yuan, Tianyuan
Liu, Zhe
Zhao, Hang
Computer Vision and Pattern Recognition
Recent advancements in multi-view scene reconstruction have been significant, yet existing methods face limitations when processing streams of input images. These methods either rely on time-consuming offline optimization or are restricted to shorter sequences, hindering their applicability in real-time scenarios. In this work, we propose LONG3R (LOng sequence streaming 3D Reconstruction), a novel model designed for streaming multi-view 3D scene reconstruction over longer sequences. Our model achieves real-time processing by operating recurrently, maintaining and updating memory with each new observation. We first employ a memory gating mechanism to filter relevant memory, which, together with a new observation, is fed into a dual-source refined decoder for coarse-to-fine interaction. To effectively capture long-sequence memory, we propose a 3D spatio-temporal memory that dynamically prunes redundant spatial information while adaptively adjusting resolution along the scene. To enhance our model's performance on long sequences while maintaining training efficiency, we employ a two-stage curriculum training strategy, each stage targeting specific capabilities. Experiments demonstrate that LONG3R outperforms state-of-the-art streaming methods, particularly for longer sequences, while maintaining real-time inference speed. Project page: https://zgchen33.github.io/LONG3R/.
title LONG3R: Long Sequence Streaming 3D Reconstruction
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2507.18255