JD-SLAM: Joint camera pose estimation and moving object segmentation for simultaneous localization and mapping in dynamic scenes

Zhai, Yujia; Lu, Baoli; Li, Weijun; Xu, Jian; Ma, Shuangyi

doi:10.1177/1729881421994447

Cited by 5 publications

(4 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The advantage of mask fusion over previous dynamic SLAM systems is that it enhances the dynamic map with semantic information from many object classes in real time. This is useful for accurately identifying multiple moving targets on the battlefield [47][48][49].…”

Section: D Reconstruction Technology Based On Passive Visionmentioning

confidence: 99%

A Review of Intelligent Vision Enhancement Technology for Battlefield

et al. 2023

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

This paper reviews the application of intelligent vision enhancement technology in the battlefield environment and explores new research directions. This paper mainly introduces three parts. First, we introduce a solution for enhancing the battlefield situation and using head-mounted displays for soldier combat. Second, we summarize three core technologies supporting intelligent vision enhancement technology: 3D environment reconstruction, tracking and registration, and situational awareness. Third, we summarize three application directions of intelligent vision enhancement technology on the battlefield. Finally, the problems and challenges in the research work are proposed, including current issues such as accurate battlefield situational awareness by augmented reality technology, multiperson collaborative management and data flow, as well as challenges in future battlefield situation enhancement and perception. The future development trend of intelligent vision enhancement technology on the battlefield has been prospected. Two meanings of this paper are as follows: the first is to review the research status of intelligent vision enhancement technology from the technical level and identify the key technical points that may restrict development in the future; the second is to analyze the advantages and disadvantages of intelligent vision enhancement technology from the level of battlefield application and the roles of users. In addition, this paper proposes how to take the lead and take initiative in future wars.

show abstract

Section: D Reconstruction Technology Based On Passive Visionmentioning

confidence: 99%

A Review of Intelligent Vision Enhancement Technology for Battlefield

et al. 2023

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

show abstract

“…Zhai, Y. et al propose a visual SLAM system for lightweight dynamic scene detection. The system is mounted on the CPU and generates semantic probabilistic maps in real time to transfer the target detection network from pixel to motion clustering, and at the same time, combines with motion camera solving and 3D target segmentation to improve the accuracy and efficiency of the detection of moving objects [22].…”

Section: Introductionmentioning

confidence: 99%

Dynamic Object Detection and Tracking in Vision SLAM

Liu,

Niu,

Deng

2024

Applied Mathematics and Nonlinear Sciences

View full text Add to dashboard Cite

Addressing the trade-off between operational efficiency and localization accuracy in visual SLAM, this paper introduces a monocular visual-inertial SLAM algorithm that integrates point and line features. To construct the point-line reprojection error and optimize the observation volume in front-end vision initialization, the motion recovery structure method (SFM) is employed through 3D reconstruction with a sliding window. The marginalization method uses the removed keyframe information as a priori constraint for nonlinear optimization in the back-end. In addition, the loopback detection algorithm is optimized in combination with the bag-of-words model and four-degree-of-freedom global bitmap to improve the accuracy of dynamic object detection, and the performance of the algorithm is tested. The results show that in the case of no closed loop, the absolute root mean square error of the algorithm proposed in this paper is lower than that of VINS-Mono (0.0625), PL-VIO (0.0401), and PL-VINS (0.0554) for the majority of sequences. In the case of closed loops, the absolute root mean square error of the proposed algorithm in this paper is reduced by 0.0395 and 0.0139 on average over most sequences compared to VINS-Mono and PL-VINS. The proposed algorithm in this paper demonstrates higher accuracy and robustness for improved detection and tracking of dynamic objects.

show abstract

“…Recently, visual odometry (VO) 3 –6 based on deep learning has achieved superior results over classical VO in some three-dimensional tasks. These methods provide new solutions for robot pose estimation tasks using point cloud data.…”

Section: Introductionmentioning

confidence: 99%

Self-supervised learning of LiDAR odometry based on spherical projection

Liu

Zhang

et al. 2022

International Journal of Advanced Robotic Systems

View full text Add to dashboard Cite

Recently, the learning-based LiDAR odometry has obtained robust estimation results in the field of mobile robot localization, but most of them are constructed based on the idea of supervised learning. In the network training stage, these supervised learning-based methods rely heavily on real pose labels, which is defective in practical applications. Different from these methods, a novel self-supervised LiDAR odometry, namely SSLO, is proposed in this article. The proposed SSLO only uses unlabeled point cloud data to train the three-view pose network to complete the robot localization task. Specifically, first, due to the sparseness and disorder of the original LiDAR point cloud, it is difficult to use deep convolutional neural networks for feature extraction. In this article, the spherical projection of the point cloud is used to convert the original point cloud into a regular vertex map. Then the vertex map obtained by the projection is used as the input of the neural network. Second, in the network training phase, SSLO uses multiple geometric losses for different situations of matching point clouds and introduces uncertainty weights when calculating the losses to reduce the interference of noise or moving objects in the scene. Last but not least, the proposed method is not only used in the simulation experiments based on the KITTI dataset and Apollo-SouthBay dataset but also applied to a real-world wheeled robot SLAM task. Extensive experimental results show that the proposed method has good performance in different environments.

show abstract

JD-SLAM: Joint camera pose estimation and moving object segmentation for simultaneous localization and mapping in dynamic scenes

Cited by 5 publications

References 20 publications

A Review of Intelligent Vision Enhancement Technology for Battlefield

A Review of Intelligent Vision Enhancement Technology for Battlefield

Dynamic Object Detection and Tracking in Vision SLAM

Self-supervised learning of LiDAR odometry based on spherical projection

Contact Info

Product

Resources

About