SLAM-Former: Putting SLAM into One Transformer

Yijun Yuan, Zhuoguang Chen, Kenan Li, Weibang Wang, Hang Zhao
9/21/2025
cs.CVcs.RO

No official code found. Best implementation: Tsinghua-MARS-Lab/SLAM-Former (511 stars, BSD-3-Clause, last commit 2026-04-25).

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.

View on arXivView PDF

Code Implementations(1)

Tsinghua-MARS-Lab/SLAM-FormerCommunity66%
BSD-3-Clause

SLAM-Former: Putting SLAM into One Transformer

51113C++, CudaSep 23, 20255 months agoBSD-3-Clause

Similar Papers

CTA-Net: A CNN-Transformer Aggregation Network for Improving Multi-Scale Feature Extraction

84% match2024

DDCL-INCRT: A Self-Organising Transformer with Hierarchical Prototype Structure (Theoretical Foundations)

84% match2026

The Progression of Transformers from Language to Vision to MOT: A Literature Review on Multi-Object Tracking with Transformers

84% match2024

Co-Me: Confidence-Guided Token Merging for Visual Geometric Transformers

84% match1 repo1962025

CTRL-F: Pairing Convolution with Transformer for Image Classification via Multi-Level Feature Cross-Attention and Representation Learning Fusion

83% match1 repo52024

Dynamic Transformer Architecture for Continual Learning of Multimodal Tasks

83% match2024

Cite this paper

@article{yuan2025slamformer,
  title  = {SLAM-Former: Putting SLAM into One Transformer},
  author = {Yijun Yuan and Zhuoguang Chen and Kenan Li and Weibang Wang and Hang Zhao},
  year   = {2025},
  eprint = {2509.16909},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2509.16909v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2509.16909.svg)](https://codeofpaper.com/paper/2509.16909)

Discussion