Awesome
DurLAR: A High-Fidelity 128-Channel LiDAR Dataset
<!-- ![DurLAR](https://github.com/l1997i/DurLAR/blob/main/head.png?raw=true) -->https://github.com/l1997i/DurLAR/assets/35445094/2c6d4056-a6de-4fad-9576-693efe2860f0
News
- [2024/09/07] We provide the OneDrive link for downloading the DurLAR dataset. To access the OneDrive link, please email your affiliation, name, and intended usage to i@luisli.org.
Sensor placement
-
LiDAR: Ouster OS1-128 LiDAR sensor with 128 channels vertical resolution
-
Stereo Camera: Carnegie Robotics MultiSense S21 stereo camera with grayscale, colour, and IR enhanced imagers, 2048x1088 @ 2MP resolution
-
GNSS/INS: OxTS RT3000v3 global navigation satellite and inertial navigation system, supporting localization from GPS, GLONASS, BeiDou, Galileo, PPP and SBAS constellations
-
Lux Meter: Yocto Light V3, a USB ambient light sensor (lux meter), measuring ambient light up to 100,000 lux
Panoramic Imagery
<br> <p align="center"> <img src="https://github.com/l1997i/DurLAR/blob/main/reflect_center.gif?raw=true" width="100%"/> <h5 id="title" align="center">Reflectivity imagery</h5> </br> <br> <p align="center"> <img src="https://github.com/l1997i/DurLAR/blob/main/ambient_center.gif?raw=true" width="100%"/> <h5 id="title" align="center">Ambient imagery</h5> </br>File Description
Each file contains 8 topics for each frame in DurLAR dataset,
ambient/
: panoramic ambient imageryreflec/
: panoramic reflectivity imageryimage_01/
: right camera (grayscale+synced+rectified)image_02/
: left RGB camera (synced+rectified)ouster_points
: ouster LiDAR point cloud (KITTI-compatible binary format)gps
,imu
,lux
: csv file format
The structure of the provided DurLAR full dataset zip file,
DurLAR_<date>/
├── ambient/
│ ├── data/
│ │ └── <frame_number.png> [ ..... ]
│ └── timestamp.txt
├── gps/
│ └── data.csv
├── image_01/
│ ├── data/
│ │ └── <frame_number.png> [ ..... ]
│ └── timestamp.txt
├── image_02/
│ ├── data/
│ │ └── <frame_number.png> [ ..... ]
│ └── timestamp.txt
├── imu/
│ └── data.csv
├── lux/
│ └── data.csv
├── ouster_points/
│ ├── data/
│ │ └── <frame_number.bin> [ ..... ]
│ └── timestamp.txt
├── reflec/
│ ├── data/
│ │ └── <frame_number.png> [ ..... ]
│ └── timestamp.txt
└── readme.md [ this README file ]
The structure of the provided calibration zip file,
DurLAR_calibs/
├── calib_cam_to_cam.txt [ Camera to camera calibration results ]
├── calib_imu_to_lidar.txt [ IMU to LiDAR calibration results ]
└── calib_lidar_to_cam.txt [ LiDAR to camera calibration results ]
Get Started
- Download the calibration files
- Download the calibration files (v2, targetless)
- Download the exemplar ROS bag (for targetless calibration)
- Download the exemplar dataset (600 frames)
- Download the full dataset (Fill in the form to request access to the full dataset)
Note that we did not include CSV header information in the exemplar dataset (600 frames). You can refer to Header of csv files to get the first line of the
csv
files.
calibration files (v2, targetless): Following the publication of the proposed DurLAR dataset and the corresponding paper, we identify a more advanced targetless calibration method (#4) that surpasses the LiDAR-camera calibration technique previously employed. We provide exemplar ROS bag for targetless calibration, and also corresponding calibration results (v2). Please refer to Appendix (arXiv) for more details.
Access to the full dataset
Access to the complete DurLAR dataset can be requested through one of the following ways. 您可任选以下其中任意链接申请访问完整数据集。
1. Access for the full dataset
3. OneDrive link. If you need access to the DurLAR dataset, please send your affiliation, name, and intended usage to i@luisli.org, and we will provide the access password.
Usage of the downloading script
Upon completion of the form, the download script durlar_download
and accompanying instructions will be automatically provided. The DurLAR dataset can then be downloaded via the command line.
For the first use, it is highly likely that the durlar_download
file will need to be made
executable:
chmod +x durlar_download
By default, this script downloads the small subset for simple testing. Use the following command:
./durlar_download
It is also possible to select and download various test drives:
usage: ./durlar_download [dataset_sample_size] [drive]
dataset_sample_size = [ small | medium | full ]
drive = 1 ... 5
Given the substantial size of the DurLAR dataset, please download the complete dataset only when necessary:
./durlar_download full 5
Throughout the entire download process, it is important that your network remains stable and free from any interruptions. In the event of network issues, please delete all DurLAR dataset folders and rerun the download script. Currently, our script supports only Ubuntu (tested on Ubuntu 18.04 and Ubuntu 20.04, amd64). For downloading the DurLAR dataset on other operating systems, please refer to Durham Collections for instructions.
CSV format for imu
, gps
, and lux
topics
Format description
Our imu
, gps
, and lux
data are all in CSV
format. The first row of the CSV
file contains headers that describe the meaning of each column. Taking imu
csv file for example (only the first 9 rows are displayed),
%time
: Timestamps in Unix epoch format.field.header.seq
: Sequence numbers.field.header.stamp
: Header timestamps.field.header.frame_id
: Frame of reference, labeled as "gps".field.orientation.x
: X-component of the orientation quaternion.field.orientation.y
: Y-component of the orientation quaternion.field.orientation.z
: Z-component of the orientation quaternion.field.orientation.w
: W-component of the orientation quaternion.field.orientation_covariance0
: Covariance of the orientation data.
Header of csv
files
The first line of the csv
files is shown as follows.
For the GPS,
time,field.header.seq,field.header.stamp,field.header.frame_id,field.status.status,field.status.service,field.latitude,field.longitude,field.altitude,field.position_covariance0,field.position_covariance1,field.position_covariance2,field.position_covariance3,field.position_covariance4,field.position_covariance5,field.position_covariance6,field.position_covariance7,field.position_covariance8,field.position_covariance_type
For the IMU,
time,field.header.seq,field.header.stamp,field.header.frame_id,field.orientation.x,field.orientation.y,field.orientation.z,field.orientation.w,field.orientation_covariance0,field.orientation_covariance1,field.orientation_covariance2,field.orientation_covariance3,field.orientation_covariance4,field.orientation_covariance5,field.orientation_covariance6,field.orientation_covariance7,field.orientation_covariance8,field.angular_velocity.x,field.angular_velocity.y,field.angular_velocity.z,field.angular_velocity_covariance0,field.angular_velocity_covariance1,field.angular_velocity_covariance2,field.angular_velocity_covariance3,field.angular_velocity_covariance4,field.angular_velocity_covariance5,field.angular_velocity_covariance6,field.angular_velocity_covariance7,field.angular_velocity_covariance8,field.linear_acceleration.x,field.linear_acceleration.y,field.linear_acceleration.z,field.linear_acceleration_covariance0,field.linear_acceleration_covariance1,field.linear_acceleration_covariance2,field.linear_acceleration_covariance3,field.linear_acceleration_covariance4,field.linear_acceleration_covariance5,field.linear_acceleration_covariance6,field.linear_acceleration_covariance7,field.linear_acceleration_covariance8
For the LUX,
time,field.header.seq,field.header.stamp,field.header.frame_id,field.illuminance,field.variance
To process the csv
files
To process the csv
files, you can use multiple ways. For example,
Python: Use the pandas library to read the CSV file with the following code:
import pandas as pd
df = pd.read_csv('data.csv')
print(df)
Text Editors: Simple text editors like Notepad
(Windows) or TextEdit
(Mac) can also open CSV
files, though they are less suited for data analysis.
Folder #Frame Verification
For easy verification of folder data and integrity, we provide the number of frames in each drive folder, as well as the MD5 checksums of the zip files.
Folder | # of Frames |
---|---|
20210716 | 41993 |
20210901 | 23347 |
20211012 | 28642 |
20211208 | 26850 |
20211209 | 25079 |
total | 145911 |
Reference
If you are making use of this work in any way (including our dataset and toolkits), you must please reference the following paper in any report, publication, presentation, software release or any other associated materials:
DurLAR: A High-fidelity 128-channel LiDAR Dataset with Panoramic Ambient and Reflectivity Imagery for Multi-modal Autonomous Driving Applications (Li Li, Khalid N. Ismail, Hubert P. H. Shum and Toby P. Breckon), In Int. Conf. 3D Vision, 2021. [pdf] [video][poster]
@inproceedings{li21durlar,
author = {Li, L. and Ismail, K.N. and Shum, H.P.H. and Breckon, T.P.},
title = {DurLAR: A High-fidelity 128-channel LiDAR Dataset with Panoramic Ambient and Reflectivity Imagery for Multi-modal Autonomous Driving Applications},
booktitle = {Proc. Int. Conf. on 3D Vision},
year = {2021},
month = {December},
publisher = {IEEE},
keywords = {autonomous driving, dataset, high-resolution LiDAR, flash LiDAR, ground truth depth, dense depth, monocular depth estimation, stereo vision, 3D},
category = {automotive 3Dvision},
}