SLAM3R：基于单目RGB视频的实时稠密场景重建
SLAM3R: Real-Time Dense Scene Reconstruction from Monocular RGB Videos
AI总结：
针对单目RGB视频实时稠密三维重建需求，提出SLAM3R系统，通过前馈网络整合局部重建与全局配准，无需显式求解相机参数，在多数据集上实现SOTA精度且帧率超20FPS。
摘要：
本文提出了SLAM3R，这是一种新颖且有效的系统，可利用RGB视频实现实时、高质量的稠密三维重建。SLAM3R通过前馈神经网络无缝整合局部三维重建与全局坐标配准，提供了一套端到端解决方案。系统接收输入视频后，首先通过滑动窗口机制将其转换为重叠的视频片段。与传统基于位姿优化的方法不同，SLAM3R直接从每个窗口内的RGB图像回归三维点图，并逐步对齐、变形这些局部点图以生成全局一致的场景重建结果——全程无需显式求解任何相机参数。在多个数据集上的实验一致表明，SLAM3R达到了当前最优的重建精度与完整度，同时保持了20FPS以上的实时性能。代码可在以下地址获取：https://github.com/PKU-VCL-3DV/SLAM3R。
英文摘要：
In this paper, we introduce SLAM3R, a novel and effective system for real-time, high-quality, dense 3D reconstruction using RGB videos. SLAM3R provides an end-to-end solution by seamlessly integrating local 3D reconstruction and global coordinate registration through feed-forward neural networks. Given an input video, the system first converts it into overlapping clips using a sliding window mechanism. Unlike traditional pose optimization-based methods, SLAM3R directly regresses 3D pointmaps from RGB images in each window and progressively aligns and deforms these local pointmaps to create a globally consistent scene reconstruction - all without explicitly solving any camera parameters. Experiments across datasets consistently show that SLAM3R achieves state-of-the-art reconstruction accuracy and completeness while maintaining real-time performance at 20+ FPS. Code available at: https://github.com/PKU-VCL-3DV/SLAM3R.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
