PaperSwipe

SLAM-Former: Putting SLAM into One Transformer

Published 2 months agoVersion 1arXiv:2509.16909

Authors

Yijun Yuan, Zhuoguang Chen, Kenan Li, Weibang Wang, Hang Zhao

Categories

cs.CVcs.RO

Abstract

We present SLAM-Former, a novel neural approach that integrates full SLAM capabilities into a single transformer. Similar to traditional SLAM systems, SLAM-Former comprises both a frontend and a backend that operate in tandem. The frontend processes sequential monocular images in real-time for incremental mapping and tracking, while the backend performs global refinement to ensure a geometrically consistent result. This alternating execution allows the frontend and backend to mutually promote one another, enhancing overall system performance. Comprehensive experimental results demonstrate that SLAM-Former achieves superior or highly competitive performance compared to state-of-the-art dense SLAM methods.

SLAM-Former: Putting SLAM into One Transformer

2 months ago
v1
5 authors

Categories

cs.CVcs.RO

Abstract

We present SLAM-Former, a novel neural approach that integrates full SLAM capabilities into a single transformer. Similar to traditional SLAM systems, SLAM-Former comprises both a frontend and a backend that operate in tandem. The frontend processes sequential monocular images in real-time for incremental mapping and tracking, while the backend performs global refinement to ensure a geometrically consistent result. This alternating execution allows the frontend and backend to mutually promote one another, enhancing overall system performance. Comprehensive experimental results demonstrate that SLAM-Former achieves superior or highly competitive performance compared to state-of-the-art dense SLAM methods.

Authors

Yijun Yuan, Zhuoguang Chen, Kenan Li et al. (+2 more)

arXiv ID: 2509.16909
Published Sep 21, 2025

Click to preview the PDF directly in your browser