Semantic3D

Dataset Information
Modalities
Images, 3D, Point cloud
Introduced
2017
License
CC BY-NC-SA 3.0
Homepage
Official Website
Contents

Overview

Semantic3D is a point cloud dataset of scanned outdoor scenes with over 3 billion points. It contains 15 training and 15 test scenes annotated with 8 class labels. This large labelled 3D point cloud data set of natural covers a range of diverse urban scenes: churches, streets, railroad tracks, squares, villages, soccer fields, castles to name just a few. The point clouds provided are scanned statically with state-of-the-art equipment and contain very fine details.

Source: Tangent Convolutions for Dense Prediction in 3D
Image Source: http://www.semantic3d.net/

Variants: Semantic3D

Associated Benchmarks

This dataset is used in 1 benchmark:

Recent Benchmark Submissions

Task Model Paper Date
Semantic Segmentation RFCR Omni-supervised Point Cloud Segmentation via … 2021-05-21
Semantic Segmentation BAAF-Net Semantic Segmentation for Real Point … 2021-03-12
Semantic Segmentation Feature Geometric Net FG-Net: Fast Large-Scale LiDAR Point … 2020-12-17
Semantic Segmentation RandLA-Net RandLA-Net: Efficient Semantic Segmentation of … 2019-11-25
Semantic Segmentation shellnet_v2 ShellNet: Efficient Point Cloud Convolutional … 2019-08-17
Semantic Segmentation KPConv KPConv: Flexible and Deformable Convolution … 2019-04-18
Semantic Segmentation RF_MSSF Semantic Classification of 3D Point … 2018-08-01
Semantic Segmentation MSDeepVoxNet Classification of Point Cloud Scenes … 2018-04-10
Semantic Segmentation SPG Large-scale Point Cloud Semantic Segmentation … 2017-11-27
Semantic Segmentation SPG Large-scale Point Cloud Semantic Segmentation … 2017-11-27
Semantic Segmentation SegCloud SEGCloud: Semantic Segmentation of 3D … 2017-10-20
Semantic Segmentation 3D-FCNN-TI SEGCloud: Semantic Segmentation of 3D … 2017-10-20
Semantic Segmentation DeePr3SS Deep Projective 3D Semantic Segmentation 2017-05-09

Research Papers

Recent papers with results on this dataset:

External Links: