Need help with kitti360Scripts?
Click the “chat” button below for chat support from the developer who created it, or find similar developers for support.

About the developer

182 Stars 29 Forks Other 18 Commits 15 Opened issues

Services available


Need anything else?

Contributors list

The KITTI-360 Dataset

This repository contains scripts for inspection of the KITTI-360 dataset. This large-scale dataset contains 320k images and 100k laser scans in a driving distance of 73.7km. We annotate both static and dynamic 3D scene elements with rough bounding primitives and transfer this information into the image domain, resulting in dense semantic & instance annotations on both 3D point clouds and 2D images.

Details and download are available at:

Dataset Structure

Dataset structure and data formats are available at:




pip install git+

For the 2D graphical tools you additionally need to install

sudo apt install python-tk python-qt5

We use open3D to visualize 3D point clouds and 3D bounding boxes:

pip install open3d


This scripts contains helpers for loading and visualizing our dataset. For inspection, please download the dataset and add the root directory to your system path at first:


You can inspect the 2D images and labels using the following tool:

cd kitti360scripts/viewer

You can visualize the 3D fused point clouds and labels using the following tool:

cd kitti360scripts/viewer
python -sequence ${sequence}

Package Content

The package is structured as follows -

: helper files that are included by other scripts -
: view the 2D image & labels and the 3D fused point clouds & labels

Note that all files have a small documentation at the top. Most important files -

: central file defining the IDs of all semantic classes and providing mapping between various class properties. -
: central file containing the loaders for all labels including 3D bounding boxes and fused point clouds


The 2D graphical tool is adapted from Cityscapes.


If you find this code or our dataset helpful in your research, please use the following BibTeX entry.

   title   = {{KITTI}-360: A Novel Dataset and Benchmarks for Urban Scene Understanding in 2D and 3D}, 
   author  = {Yiyi Liao and Jun Xie and Andreas Geiger}, 
   journal = {},
   volume  = {2109.13410},
   year    = {2021}, 


Please feel free to contact us with any questions, suggestions or comments:

We use cookies. If you continue to browse the site, you agree to the use of cookies. For more information on our use of cookies please see our Privacy Policy.