English

vMAP: Vectorised Object Mapping for Neural Field SLAM

Computer Vision and Pattern Recognition 2023-03-15 v2

Abstract

We present vMAP, an object-level dense SLAM system using neural field representations. Each object is represented by a small MLP, enabling efficient, watertight object modelling without the need for 3D priors. As an RGB-D camera browses a scene with no prior information, vMAP detects object instances on-the-fly, and dynamically adds them to its map. Specifically, thanks to the power of vectorised training, vMAP can optimise as many as 50 individual objects in a single scene, with an extremely efficient training speed of 5Hz map update. We experimentally demonstrate significantly improved scene-level and object-level reconstruction quality compared to prior neural field SLAM systems. Project page: https://kxhit.github.io/vMAP.

Keywords

Cite

@article{arxiv.2302.01838,
  title  = {vMAP: Vectorised Object Mapping for Neural Field SLAM},
  author = {Xin Kong and Shikun Liu and Marwan Taher and Andrew J. Davison},
  journal= {arXiv preprint arXiv:2302.01838},
  year   = {2023}
}

Comments

CVPR2023 Project Page:https://kxhit.github.io/vMAP

R2 v1 2026-06-28T08:31:30.649Z