Efficient 3D semantic segmentation with superpoint transformer
We introduce a novel superpoint-based transformer architecture for efficient semantic
segmentation of large-scale 3D scenes. Our method incorporates a fast algorithm to partition …
segmentation of large-scale 3D scenes. Our method incorporates a fast algorithm to partition …
Self-supervised 3d scene flow estimation guided by superpoints
Abstract 3D scene flow estimation aims to estimate point-wise motions between two
consecutive frames of point clouds. Superpoints, ie, points with similar geometric features …
consecutive frames of point clouds. Superpoints, ie, points with similar geometric features …
Point cloud oversegmentation with graph-structured deep metric learning
We propose a new supervized learning framework for oversegmenting 3D point clouds into
superpoints. We cast this problem as learning deep embeddings of the local geometry and …
superpoints. We cast this problem as learning deep embeddings of the local geometry and …
Rigidflow: Self-supervised scene flow learning on point clouds by local rigidity prior
In this work, we focus on scene flow learning on point clouds in a self-supervised manner. A
real-world scene can be well modeled as a collection of rigidly moving parts, therefore its …
real-world scene can be well modeled as a collection of rigidly moving parts, therefore its …
Context-aware network for semantic segmentation toward large-scale point clouds in urban environments
C Liu, D Zeng, A Akbar, H Wu, S Jia… - IEEE Transactions on …, 2022 - ieeexplore.ieee.org
Point cloud semantic segmentation in urban scenes plays a vital role in intelligent city
modeling, autonomous driving, and urban planning. Point cloud semantic segmentation …
modeling, autonomous driving, and urban planning. Point cloud semantic segmentation …
Unsupervised reconstruction of Building Information Modeling wall objects from point cloud data
Scan-to-BIM of existing buildings is in high demand by the construction industry. However,
these models are costly and time-consuming to create. The automation of this process is still …
these models are costly and time-consuming to create. The automation of this process is still …
Multi-scale point-wise convolutional neural networks for 3D object segmentation from LiDAR point clouds in large-scale environments
Although significant improvement has been achieved in fully autonomous driving and
semantic high-definition map (HD) domains, most of the existing 3D point cloud …
semantic high-definition map (HD) domains, most of the existing 3D point cloud …
Complete-to-partial 4D distillation for self-supervised point cloud sequence representation learning
Recent work on 4D point cloud sequences has attracted a lot of attention. However,
obtaining exhaustively labeled 4D datasets is often very expensive and laborious, so it is …
obtaining exhaustively labeled 4D datasets is often very expensive and laborious, so it is …