The Perfect Match: 3D Point Cloud Matching with Smoothed Densities

11/16/2018
by   Zan Gojcic, et al.
6

We propose 3DSmoothNet, a full workflow to match 3D point clouds with a siamese deep learning architecture and fully convolutional layers using a voxelized smoothed density value (SDV) representation. The latter is computed per interest point and aligned to the local reference frame (LRF) to achieve rotation invariance. Our compact, learned, rotation invariant 3D point cloud descriptor achieves 94.9 outperforming the state-of-the-art by more than 20 percent points with only 32 output dimensions. This very low output dimension allows for near real-time correspondence search with 0.1 ms per feature point on a standard PC. Our approach is sensor- and scene-agnostic because of SDV, LRF and learning highly descriptive features with fully convolutional layers. We show that 3DSmoothNet trained only on RGB-D indoor scenes of buildings achieves 79.0 on laser scans of outdoor vegetation, more than double the performance of our closest, learningbased competitors.

READ FULL TEXT

page 3

page 4

page 5

page 6

01/02/2021

Learning Rotation-Invariant Representations of Point Clouds Using Aligned Edge Convolutional Neural Networks

Point cloud analysis is an area of increasing interest due to the develo...
05/21/2021

Generalisable and distinctive 3D local deep descriptors for point cloud registration

An effective 3D descriptor should be invariant to different geometric tr...
09/01/2020

Distinctive 3D local deep descriptors

We present a simple but yet effective method for learning distinctive 3D...
07/06/2020

Learning Graph-Convolutional Representations for Point Cloud Denoising

Point clouds are an increasingly relevant data type but they are often c...
09/01/2021

You Only Hypothesize Once: Point Cloud Registration with Rotation-equivariant Descriptors

In this paper, we propose a novel local descriptor-based framework, call...
08/30/2018

PPF-FoldNet: Unsupervised Learning of Rotation Invariant 3D Local Descriptors

We present PPF-FoldNet for unsupervised learning of 3D local descriptors...
03/22/2018

Aligning Across Large Gaps in Time

We present a method of temporally-invariant image registration for outdo...

Code Repositories

3DSmoothNet

[CVPR2019] The Perfect Match: 3D Point Cloud Matching with Smoothed Densities


view repo

Innav

Что я делаю с проектом Innav


view repo