Skip to content

Datasets used in photogrammetry, computer vision, laser scanning, 3D modelling

Notifications You must be signed in to change notification settings

mvarga1989/Datasets_Reality_capture_3Dmodelling

Folders and files

NameName
Last commit message
Last commit date

Latest commit

ย 

History

14 Commits
ย 
ย 

Repository files navigation

You wish to assist in making the list more useful, complete and accurate? Contributors are more than welcome. Please send me an ๐Ÿ“ง.

If this content has added any value to your work, please consider giving it a โญ and/or share on your professional social networks. Thank you!

Other useful lists that I maintain are available here ๐Ÿ”—

Find out more about me here ๐Ÿ”— and here ๐Ÿ”—

Photogrammetry, LiDAR, laser scanning data collections

Websites where you can download the Lidar datasets for free

  • USGS Earth Explorer
  • NOAA Digital Coast
  • Open Topography
  • National Ecological Observatory Network (NEON).

Individual datasets

  • Omni Warsaw Dataset
  • UseGeo - UAV-based multi-sensor dataset for geospatial research
    • a large-scale real-world set of data suitable for multiple tasks, such as monocular depth estimation, multi-view 3D reconstruction (MVS), etc.
  • LiSurveying: A high-resolution TLS-LiDAR benchmark
    • The first large-scale hybrid (point-wise and 3D bounding box) TLS-LiDAR dataset for LS.
    • LiSurveying provides 360 degrees of sensor coverage in multiple locations of the scene, where multiple scans are integrated to obtain a massive point-cloud representation.
    • LiSurveying provides rich data for research on object classification, semantic segmentation, and object detection for complex outdoor scenes.
  • An open-source photogrammetry workflow for reconstructing 3D models
  • KITTI
    • a pioneer multimodal dataset that includes point-clouds from a LiDAR sensor as well as front-facing stereo images and GPS/IMU data. This dataset with 8 classes is useful for AV applications using MLS point-clouds and front-facing stereo images. This dataset contains about 15k point-cloud frames with 200k 3D bounding boxes on 22 different scenes.
  • Paris-Lille-3D
  • Toronto-3D

Geomatics

  • Landslide4Sense
    • Reference Benchmark Data and Deep Learning Models for Landslide Detection

Useful

Releases

No releases published

Packages

No packages published