Automotive
DurLAR
depth estimation; multimodal driving
Repeated Durham driving routes with high-resolution LiDAR, stereo and LiDAR-derived ambient/reflectivity images.
Access links
Primary places to inspect, download, read about, or reproduce the dataset.
Quick facts
- Vehicle type
- Automotive
- Environment
- Outdoor
- Data origin
- Real-world
- Sensor count
- 5
- Ground truth
- Available
- Calibration
- Partial Raw Data
- Annotations
- Available
Sensor overview
High-level sensor availability before the detailed sensor records below.
Ground truth
Reference scope, method, rate, coverage, and provenance.
| Scope | Method | Reference system | Count | Rate | Coverage |
Reference provenance
|
|---|---|---|---|---|---|---|
| Other | Other | Projected OS1-128 range | Not reported | Not reported | Unknown |
Reference provenance
Derived
|
|
||||||
Calibration and synchronization
Reported calibration level, reproducibility signals, and supporting notes.
- Level
- Partial Raw Data
- Processed parameters
- Available
- Raw calibration data
- Available
- Calibration targets
- Not reported
Reproducible calibration data
Raw calibration data is reported, so users can reproduce or inspect the calibration process.
Notes
Original/v2 parameter archives, LiDAR intrinsic JSON and an exemplar raw bag for targetless calibration are documented. This is partial raw material, not complete raw sessions for every sensor.
Known or intentional limitations
Reported constraints and characteristics to check before using the dataset.
Five drive folders contain 145,911 frames in the maintained manifest. The original 600-frame sample differs from the full dataset; current full access uses gated Hugging Face authentication and author approval.
Camera maximum frame rates, GNSS/INS logging rates and native component sampling have different meanings. Mixed RGB/grayscale stereo, sparse projected depth and exact per-channel cadence need timestamp-aware handling.
Ouster spinning acquisition conflicts with flash/no-skew prose; wavelength claims also differ between paper sections. No undistorted-motion or universal sparse-depth accuracy claim is made.
One physical LiDAR also supplies ambient/reflectivity imagery. Exemplar raw calibration data do not establish full reproducibility for all devices or independently measured pose/depth/timing accuracy.
Cameras
Camera 1
- Model
- Carnegie Robotics MultiSense S21
- Setup
- Stereo
- Count
- 1
- Effective cameras
- 2
- Modality
- Rgb
- Resolution
- 2048 x 1088
- Rate
- Not reported
- Shutter
- Not Reported
- Lens type
- Not Reported
- HFOV
- Not reported
- VFOV
- Not reported
- Notes
- RGB is a coarse availability label. One MultiSense S21 stereo system supplies left RGB and right grayscale images at 2048x1088. Camera hardware maximum 30 fps and 10 Hz cross-sensor pairing do not establish the native delivered camera cadence.
IMUs
OxTS RT3000v3 inertial component
- Model
- Not reported
- Count
- 1
- Accel / gyro
- Not reported
GNSS
OxTS RT3000v3
- Model or name
- OxTS RT3000v3
- Count
- 1
- Rate
- Not reported
- GNSS type
- GPS_IMU
- Position output
- Available
- Velocity output
- Not reported
- Heading output
- Not reported
LiDAR
Ouster OS1-128
- Model or name
- Ouster OS1-128
- Count
- 1
- Dimensions
- 3
- Rate
- 10 Hz
- Channels
- 128
- Range
- Not reported
Radar
Not reported
Additional sensors
Other sensor
- Model
- Yocto Light V3
- Name
- Ambient illuminance
- Count
- 1
- Notes
- Timestamped illuminance and variance CSV from Yocto Light V3. The 100,000 lux device limit is not an observed dataset maximum; native cadence remains unknown.
Annotations
Reported annotation availability and task support.
- Availability
- Available
- Format
- Not reported
- Class count
- Not reported
Semantic segmentation
Not reported
Instance segmentation
Not reported
Object detection
Not reported
Optical flow
Not reported
Depth ground truth
Available
Citation
Citation key and BibTeX kept at the end of the page for reference.
@inproceedings{li2021durlar,
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},
}Dataset corrections
Notice a specific issue after reviewing the page?
Report an inaccuracy