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

Wu, Yanmin; Zhang, Yunzhou; Zhu, Delong; Feng, Yonghui; Coleman, Sonya; Kerr, Dermot

doi:10.1109/iros45743.2020.9341757

Cited by 67 publications

(47 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Accurate object representation is a key issue in objectoriented SLAM research and 3D object models [14], cubic boxes [3]- [5] and ellipsoids [6]- [8] are among common methods utilized for object representation. Prior work like [4] and [5] use the cubic box to represent the object, where the pose of the cubic box can be estimated by vanishing Fig. 1.…”

Section: Introductionmentioning

confidence: 99%

Accurate and Robust Object SLAM With 3D Quadric Landmark Reconstruction in Outdoors

Tian

Zhang

Feng

et al. 2022

IEEE Robot. Autom. Lett.

Self Cite

View full text Add to dashboard Cite

Object-oriented SLAM is a popular technology in autonomous driving and robotics. In this paper, we propose a stereo visual SLAM with a robust quadric landmark representation method. The system consists of four components, including deep learning detection, quadric landmark initialization, object data association and object pose optimization. State-of-the-art quadric-based SLAM algorithms always face observation related problems and are sensitive to observation noise, which limits their application in outdoor scenes. To solve this problem, we propose a quadric initialization method based on the separation of the quadric parameters method, which improves the robustness to observation noise. The sufficient object data association algorithm and object-oriented optimization with multiple cues enables a highly accurate object pose estimation that is robust to local observations. Experimental results show that the proposed system is more robust to observation noise and significantly outperforms current stateof-the-art methods in outdoor environments. In addition, the proposed system demonstrates real-time performance.

show abstract

Section: Introductionmentioning

confidence: 99%

Accurate and Robust Object SLAM With 3D Quadric Landmark Reconstruction in Outdoors

Tian

Zhang

Feng

et al. 2022

IEEE Robot. Autom. Lett.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Ref. [248] proposed a lightweight object-oriented SLAM system, which effectively solves the problems of data association and attitude estimation, and solves the problem of the poor real-time performance of the above methods. The core framework is developed based on ORB-SLAM2 and uses YOLOv3 as an object detector to fuse semantic thread.…”

Section: For Sencementioning

confidence: 99%

An Overview on Visual SLAM: From Tradition to Semantic

et al. 2022

View full text Add to dashboard Cite

Visual SLAM (VSLAM) has been developing rapidly due to its advantages of low-cost sensors, the easy fusion of other sensors, and richer environmental information. Traditional visionbased SLAM research has made many achievements, but it may fail to achieve wished results in challenging environments. Deep learning has promoted the development of computer vision, and the combination of deep learning and SLAM has attracted more and more attention. Semantic information, as high-level environmental information, can enable robots to better understand the surrounding environment. This paper introduces the development of VSLAM technology from two aspects: traditional VSLAM and semantic VSLAM combined with deep learning. For traditional VSLAM, we summarize the advantages and disadvantages of indirect and direct methods in detail and give some classical VSLAM open-source algorithms. In addition, we focus on the development of semantic VSLAM based on deep learning. Starting with typical neural networks CNN and RNN, we summarize the improvement of neural networks for the VSLAM system in detail. Later, we focus on the help of target detection and semantic segmentation for VSLAM semantic information introduction. We believe that the development of the future intelligent era cannot be without the help of semantic technology. Introducing deep learning into the VSLAM system to provide semantic information can help robots better perceive the surrounding environment and provide people with higher-level help.

show abstract

“…However, there are background map points that do not belong to these objects. As a result, we first use the depth distribution of points to filter initial outliers and then apply the isolated forest algorithm mentioned in EAO-SLAM [15] to cluster the map points and eliminate most of the remaining outliers. Finally, by referring to the convex polyhedron construction algorithm proposed in [16], we implement a surfel construction method that uses a 3D point set to obtain the tangent planes of the object surface.…”

Section: B Tangent Planes From Surfel Constructionmentioning

confidence: 99%

Object-Aware SLAM Based on Efficient Quadric Initialization and Joint Data Association

Cao

Zhang

Tian

et al. 2022

IEEE Robot. Autom. Lett.

Self Cite

View full text Add to dashboard Cite

Semantic simultaneous localization and mapping (SLAM) is a popular technology enabling indoor mobile robots to sufficiently perceive and interact with the environment. In this paper, we propose an object-aware semantic SLAM system, which consists of a quadric initialization method, an object-level data association method, and a multi-constraint optimization factor graph. To overcome the limitation of multi-view observations and the requirement of dense point clouds for objects, an efficient quadric initialization method based on object detection and surfel construction is proposed, which can efficiently initialize quadrics within fewer frames and with small viewing angles. The robust object-level joint data association method and the tightly coupled multi-constraint factor graph for quadrics optimization and joint bundle adjustment enable the accurate estimation of constructed quadrics and camera poses. Extensive experiments using public datasets show that the proposed system achieves competitive performance with respect to accuracy and robustness of object quadric estimation and camera localization compared with stateof-the-art methods. I. INTRODUCTIONO BJECT representation is a key issue within object-level semantic SLAM, and appropriate representation can not only promote the robustness and accuracy of localization but also enhance the information of a semantic map oriented for human-robot interactions. There are many kinds of object representation methods, including preset models and general models, where prior object point clouds, cubes and quadrics are three kinds of common representation methods utilized for object-level semantic SLAM [1]-[10]. SLAM++ [1] presents an object-oriented 3D SLAM paradigm, which requires prior CAD models of objects. CubeSLAM [2] models objects as cubes, which is the first example of object-oriented SLAM. Compared with the cube methods, the quadric approach has a Manuscript

show abstract

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

Cited by 67 publications

References 26 publications

Accurate and Robust Object SLAM With 3D Quadric Landmark Reconstruction in Outdoors

Accurate and Robust Object SLAM With 3D Quadric Landmark Reconstruction in Outdoors

An Overview on Visual SLAM: From Tradition to Semantic

Object-Aware SLAM Based on Efficient Quadric Initialization and Joint Data Association

Contact Info

Product

Resources

About