EAO-SLAM: Monocular Semi-Dense Object SLAM Based on Ensemble Data Association

Abstract

Object-level data association and pose estimation play a fundamental role in semantic SLAM, which remain unsolved due to the lack of robust and accurate algorithms. In this work, we propose an ensemble data associate strategy for integrating the parametric and nonparametric statistic tests. By exploiting the nature of different statistics, our method can effectively aggregate the information of different measurements, and thus significantly improve the robustness and accuracy of data association. We then present an accurate object pose estimation framework, in which an outliers-robust centroid and scale estimation algorithm and an object pose initialization algorithm are developed to help improve the optimality of pose estimation results. Furthermore, we build a SLAM system that can generate semi-dense or lightweight object-oriented maps with a monocular camera. Extensive experiments are conducted on three publicly available datasets and a real scenario. The results show that our approach significantly outperforms state-of-the-art techniques in accuracy and robustness. The source code is available on: https://github.com/yanmin-wu/EAO-SLAM.

arXiv commentAccepted to IROS 2020. Project Page: https://yanmin-wu.github.io/project/eaoslam/; Code: https://github.com/yanmin-wu/EAO-SLAM

References in corpus (6)

• ORB-SLAM2: an Open-Source SLAM System for Monocular, Stereo and RGB-D Cameras
• CubeSLAM: Monocular 3D Object SLAM
• Volumetric Instance-Aware Semantic Mapping and 3D Object Discovery
• EM-Fusion: Dynamic Object-Level SLAM with Probabilistic Data Association
• TartanAir: A Dataset to Push the Limits of Visual SLAM
• Integrating Objects into Monocular SLAM: Line Based Category Specific Models

Cited by in corpus (12)

• Stereo CenterNet based 3D Object Detection for Autonomous Driving
• An Object SLAM Framework for Association, Mapping, and High-Level Tasks
• RO-MAP: Real-Time Multi-Object Mapping with Neural Radiance Fields
• Object SLAM-Based Active Mapping and Robotic Grasping
• Applications of Spiking Neural Networks in Visual Place Recognition
• ELLIPSDF: Joint Object Pose and Shape Optimization with a Bi-level Ellipsoid and Signed Distance Function Description
• Image Patch-Matching with Graph-Based Learning in Street Scenes
• OrcVIO: Object residual constrained Visual-Inertial Odometry
• Accurate and Robust Scale Recovery for Monocular Visual Odometry Based on Plane Geometry
• Accurate and Robust Object-oriented SLAM with 3D Quadric Landmark Construction in Outdoor Environment
• MSG-Loc: Multi-Label Likelihood-based Semantic Graph Matching for Object-Level Global Localization
• Self-supervised Learning of 3D Object Understanding by Data Association and Landmark Estimation for Image Sequence