RORNet: Partial-to-partial registration network with reliable overlap** representations
Three-dimensional point cloud registration is an important field in computer vision. Recently,
due to the increasingly complex scenes and incomplete observations, many partial-overlap …
due to the increasingly complex scenes and incomplete observations, many partial-overlap …
Dual-graph attention convolution network for 3-D point cloud classification
Three-dimensional point cloud classification is fundamental but still challenging in 3-D
vision. Existing graph-based deep learning methods fail to learn both low-level extrinsic and …
vision. Existing graph-based deep learning methods fail to learn both low-level extrinsic and …
Mvsalnet: Multi-view augmentation for rgb-d salient object detection
RGB-D salient object detection (SOD) enjoys significant advantages in understanding 3D
geometry of the scene. However, the geometry information conveyed by depth maps are …
geometry of the scene. However, the geometry information conveyed by depth maps are …
Rethinking training strategy in stereo matching
In stereo matching, various learning-based approaches have shown impressive
performance in solving traditional difficulties on multiple datasets. While most progress is …
performance in solving traditional difficulties on multiple datasets. While most progress is …
Vetaverse: A survey on the intersection of Metaverse, vehicles, and transportation systems
Since 2021, the term" Metaverse" has been the most popular one, garnering a lot of interest.
Because of its contained environment and built-in computing and networking capabilities, a …
Because of its contained environment and built-in computing and networking capabilities, a …
Discriminative correspondence estimation for unsupervised rgb-d point cloud registration
Point cloud registration is a fundamental task for estimating the rigid transformation matrix
between two point clouds, and is regarded as a prerequisite for downstream vision tasks …
between two point clouds, and is regarded as a prerequisite for downstream vision tasks …
Robust real-world point cloud registration by inlier detection
Real-world point cloud registration is challenging because of large outliers in
correspondence search. The mixture variations, such as partial overlap, noise and cross …
correspondence search. The mixture variations, such as partial overlap, noise and cross …
Self-supervised point cloud representation learning via separating mixed shapes
The manual annotation for large-scale point clouds costs a lot of time and is usually
unavailable in harsh real-world scenarios. Inspired by the great success of the pre-training …
unavailable in harsh real-world scenarios. Inspired by the great success of the pre-training …
Planeseg: Building a plug-in for boosting planar region segmentation
Existing methods in planar region segmentation suffer the problems of vague boundaries
and failure to detect small-sized regions. To address these, this study presents an end-to …
and failure to detect small-sized regions. To address these, this study presents an end-to …
An overlap estimation guided feature metric approach for real point cloud registration
F Zhang, L Zhang, T He, Y Sun, S Zhao, Y Zhang… - Computers & …, 2024 - Elsevier
Real point cloud registration, involving homologous and cross-source 3D data, poses
significant challenges such as partial overlap, high noise, density disparities, and scale …
significant challenges such as partial overlap, high noise, density disparities, and scale …