Visual SLAM Framework Based on Segmentation with the Improvement of Loop Closure Detection in Dynamic Environments

Sun, Liping; Singh, Rohan P.; Kanehiro, Fumio

doi:10.20965/jrm.2021.p1385

Cited by 4 publications

(1 citation statement)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In multi-view geometry-based odometry estimation, such as ORB-SLAM2 [17] and RTAB-Map [18], the front-end consists of sensor calibration, keypoints extraction and matching, and outlier rejection modules. Under ideal conditions, where the environment is static [19] [20] and there are no issues with texture or illumination, the accuracy of odometry estimation is satisfactory [9]. However, real-world applications require high robustness of the odometry estimation system to deal with challenging scenarios.…”

Section: A Geometry-based and Learning-based Odometry Estimationmentioning

confidence: 99%

CertainOdom: Uncertainty Weighted Multi-task Learning Model for LiDAR Odometry Estimation

Sun

Ding

Yoshiyasu

et al. 2022

2022 IEEE International Conference on Robotics and Biomimetics (ROBIO)

Self Cite

View full text Add to dashboard Cite

Multi-modal fusion of sensors is a commonly used approach to enhance the performance of odometry estimation, which is also a fundamental module for mobile robots. However, the question of how to perform fusion among different modalities in a supervised sensor fusion odometry estimation task? is still one of challenging issues remains. Some simple operations, such as element-wise summation and concatenation, are not capable of assigning adaptive attentional weights to incorporate different modalities efficiently, which make it difficult to achieve competitive odometry results. Recently, the Transformer architecture has shown potential for multi-modal fusion tasks, particularly in the domains of vision with language. In this work, we propose an end-to-end supervised Transformer-based LiDAR-Inertial fusion framework (namely TransFusionOdom) for odometry estimation. The multi-attention fusion module demonstrates different fusion approaches for homogeneous and heterogeneous modalities to address the overfitting problem that can arise from blindly increasing the complexity of the model. Additionally, to interpret the learning process of the Transformer-based multi-modal interactions, a general visualization approach is introduced to illustrate the interactions between modalities. Moreover, exhaustive ablation studies evaluate different multi-modal fusion strategies to verify the performance of the proposed fusion strategy. A synthetic multi-modal dataset is made public to validate the generalization ability of the proposed fusion strategy, which also works for other combinations of different modalities. The quantitative and qualitative odometry evaluations on the KITTI dataset verify the proposed TransFusionOdom could achieve superior performance compared with other related works.

show abstract

Section: A Geometry-based and Learning-based Odometry Estimationmentioning

confidence: 99%

CertainOdom: Uncertainty Weighted Multi-task Learning Model for LiDAR Odometry Estimation

Sun

Ding

Yoshiyasu

et al. 2022

2022 IEEE International Conference on Robotics and Biomimetics (ROBIO)

Self Cite

View full text Add to dashboard Cite

show abstract

Slam loop closure detection algorithm based on MSA-SG

Zhang,

Liu

et al. 2024

Cluster Comput

View full text Add to dashboard Cite

Object Detection and Segmentation Using Deeplabv3 Deep Neural Network for a Portable X-Ray Source Model

Rogelio¹,

Dadios²,

Vicerra³

2022

JACIII

View full text Add to dashboard Cite

The primary purpose of this research is to implement Deeplabv3 architecture’s deep neural network in detecting and segmenting portable X-ray source model parts such as body, handle, and aperture in the same color scheme scenario. Similarly, the aperture is smaller with lower resolution making deep convolutional neural networks more difficult to segment. As the input feature map diminishes as the net progresses, information about the aperture or the object on a smaller scale may be lost. It recommends using Deeplabv3 architecture to overcome this issue, as it is successful for semantic segmentation. Based on the experiment conducted, the average precision of the body, handle, and aperture of the portable X-ray source model are 91.75%, 20.41%, and 6.25%, respectively. Moreover, it indicates that detecting the “body” part has the highest average precision. In contrast, the detection of the “aperture” part has the lowest average precision. Likewise, the study found that using Deeplabv3 deep neural network architecture, detection, and segmentation of the portable X-ray source model was successful but needed improvement to increase the overall mean AP of 39.47%.

show abstract

Visual SLAM Framework Based on Segmentation with the Improvement of Loop Closure Detection in Dynamic Environments

Cited by 4 publications

References 33 publications

CertainOdom: Uncertainty Weighted Multi-task Learning Model for LiDAR Odometry Estimation

CertainOdom: Uncertainty Weighted Multi-task Learning Model for LiDAR Odometry Estimation

Slam loop closure detection algorithm based on MSA-SG

Object Detection and Segmentation Using Deeplabv3 Deep Neural Network for a Portable X-Ray Source Model

Contact Info

Product

Resources

About