A review of non-rigid transformations and learning-based 3D point cloud registration methods

S Monji-Azad, J Hesser, N Löw - ISPRS Journal of Photogrammetry and …, 2023 - Elsevier
Point cloud registration is a research field where the spatial relationship between two or
more sets of points in space is determined. Point clouds are found in multiple applications …

Coarse to fine-based image–point cloud fusion network for 3D object detection

M Hao, Z Zhang, L Li, K Dong, L Cheng, P Tiwari… - Information …, 2024 - Elsevier
Enhancing original LiDAR point cloud features with virtual points has gained widespread
attention in multimodal information fusion. However, existing methods struggle to leverage …

Overview of deep learning application on visual SLAM

S Li, D Zhang, Y **: Towards the age of spatial machine intelligence
C Chen, B Wang, CX Lu, N Trigoni… - ar** has recently attracted significant attention.
Instead of creating hand-designed algorithms through exploitation of physical models or …

Differentiable registration of images and lidar point clouds with voxelpoint-to-pixel matching

J Zhou, B Ma, W Zhang, Y Fang… - Advances in Neural …, 2023 - proceedings.neurips.cc
Cross-modality registration between 2D images captured by cameras and 3D point clouds
from LiDARs is a crucial task in computer vision and robotic. Previous methods estimate 2D …

DeepI2P: Image-to-point cloud registration via deep classification

J Li, GH Lee - Proceedings of the IEEE/CVF Conference on …, 2021 - openaccess.thecvf.com
This paper presents DeepI2P: a novel approach for cross-modality registration between an
image and a point cloud. Given an image (eg from a rgb-camera) and a general point cloud …

Deep learning for visual localization and map**: A survey

C Chen, B Wang, CX Lu, N Trigoni… - IEEE Transactions on …, 2023 - ieeexplore.ieee.org
Deep-learning-based localization and map** approaches have recently emerged as a
new research direction and receive significant attention from both industry and academia …

CorrI2P: Deep image-to-point cloud registration via dense correspondence

S Ren, Y Zeng, J Hou, X Chen - IEEE Transactions on Circuits …, 2022 - ieeexplore.ieee.org
Motivated by the intuition that the critical step of localizing a 2D image in the corresponding
3D point cloud is establishing 2D-3D correspondence between them, we propose the first …

2d3d-matr: 2d-3d matching transformer for detection-free registration between images and point clouds

M Li, Z Qin, Z Gao, R Yi, C Zhu… - Proceedings of the …, 2023 - openaccess.thecvf.com
The commonly adopted detect-then-match approach to registration finds difficulties in the
cross-modality cases due to the incompatible keypoint detection and inconsistent feature …

ATOP: An attention-to-optimization approach for automatic LiDAR-camera calibration via cross-modal object matching

Y Sun, J Li, Y Wang, X Xu, X Yang… - IEEE Transactions on …, 2022 - ieeexplore.ieee.org
Due to the difference of data modalities, it'sa very challenging task to find the feature
correspondences between 2D and 3D data in LiDAR-Camera calibration. In existing works …