arXiv 2509.16909
SLAM-Former: Putting SLAM into One Transformer
By Yijun Yuan, Zhuoguang Chen, et al.
Published 2025-09-21
Mindmap
Browse the paper's core ideas, clusters, and relationships in a structured outline.
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.…