# MRASfM: Multi-Camera Reconstruction and Aggregation through Structure-from-Motion in Driving Scenes

arXiv: https://arxiv.org/abs/2510.15467 · HTML: https://mlsanity.com/paper/2510.15467/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 22 of its 36 references that are in the map.
- First posted on arXiv 17 Oct 2025.

## Builds on (most cited first, 22 of 22)

- [SuperPoint: Self-Supervised Interest Point Detection and Description](https://mlsanity.com/paper/1712.07629/) (2017)
- [ORB-SLAM3: An Accurate Open-Source Library for Visual, Visual-Inertial and Multi-Map SLAM](https://mlsanity.com/paper/2007.11898/) (2020)
- [DROID-SLAM: Deep Visual SLAM for Monocular, Stereo, and RGB-D Cameras](https://mlsanity.com/paper/2108.10869/) (2021)
- [LightGlue: Local Feature Matching at Light Speed](https://mlsanity.com/paper/2306.13643/) (2023)
- [Global Structure-from-Motion Revisited](https://mlsanity.com/paper/2407.20219/) (2024)
- [Pixel-Perfect Structure-from-Motion with Featuremetric Refinement](https://mlsanity.com/paper/2108.08291/) (2021)
- [ASpanFormer: Detector-Free Image Matching with Adaptive Span Transformer](https://mlsanity.com/paper/2208.14201/) (2022)
- [High-quality Surface Reconstruction using Gaussian Surfels](https://mlsanity.com/paper/2404.17774/) (2024)
- [A Survey of Structure from Motion](https://mlsanity.com/paper/1701.08493/) (2017)
- [Graph-Based Parallel Large Scale Structure from Motion](https://mlsanity.com/paper/1912.10659/) (2019)
- [RoMe: Towards Large Scale Road Surface Reconstruction via Mesh Representation](https://mlsanity.com/paper/2306.11368/) (2023)
- [MV-Map: Offboard HD-Map Generation with Multi-view Consistency](https://mlsanity.com/paper/2305.08851/) (2023)
- [EMIE-MAP: Large-Scale Road Surface Reconstruction Based on Explicit Mesh and Implicit Encoding](https://mlsanity.com/paper/2403.11789/) (2024)
- [HybridOcc: NeRF Enhanced Transformer-based Multi-Camera 3D Occupancy Prediction](https://mlsanity.com/paper/2408.09104/) (2024)
- [3D Neural Edge Reconstruction](https://mlsanity.com/paper/2405.19295/) (2024)
- [BAMF-SLAM: Bundle Adjusted Multi-Fisheye Visual-Inertial SLAM Using Recurrent Field Transforms](https://mlsanity.com/paper/2306.01173/) (2023)
- [A Vision-Centric Approach for Static Map Element Annotation](https://mlsanity.com/paper/2309.11754/) (2023)
- [MAVIS: Multi-Camera Augmented Visual-Inertial SLAM using SE2(3) Based Exact IMU Pre-integration](https://mlsanity.com/paper/2309.08142/) (2023)
- [AdaSfM: From Coarse Global to Fine Incremental Adaptive Structure from Motion](https://mlsanity.com/paper/2301.12135/) (2023)
- [OCC-VO: Dense Mapping via 3D Occupancy-Based Visual Odometry for Autonomous Driving](https://mlsanity.com/paper/2309.11011/) (2023)
- [Distributed bundle adjustment with block-based sparse matrix compression for super large scale datasets](https://mlsanity.com/paper/2307.08383/) (2023)
- [CAMAv2: A Vision-Centric Approach for Static Map Element Annotation](https://mlsanity.com/paper/2407.21331/) (2024)

## Abstract (arXiv)

Structure from Motion (SfM) estimates camera poses and reconstructs point clouds, forming a foundation for various tasks. However, applying SfM to driving scenes captured by multi-camera systems presents significant difficulties, including unreliable pose estimation, excessive outliers in road surface reconstruction, and low reconstruction efficiency. To address these limitations, we propose a Multi-camera Reconstruction and Aggregation Structure-from-Motion (MRASfM) framework specifically designed for driving scenes. MRASfM enhances the reliability of camera pose estimation by leveraging the fixed spatial relationships within the multi-camera system during the registration process. To improve the quality of road surface reconstruction, our framework employs a plane model to effectively remove erroneous points from the triangulated road surface. Moreover, treating the multi-camera set as a single unit in Bundle Adjustment (BA) helps reduce optimization variables to boost efficiency. In addition, MRASfM achieves multi-scene aggregation through scene association and assembly modules in a coarse-to-fine fashion. We deployed multi-camera systems on actual vehicles to validate the generalizability of MRASfM across various scenes and its robustness in challenging conditions through real-world applications. Furthermore, large-scale validation results on public datasets show the state-of-the-art performance of MRASfM, achieving 0.124 absolute pose error on the nuScenes dataset.