mobility-datasets 0.1.2

Python library for downloading, loading, and working with autonomous driving and mobility datasets

mobility-datasets

Python library for downloading and managing autonomous driving datasets like KITTI, nuScenes, and Waymo.

---

Features

• Easy Downloads: Simple CLI and Python API for dataset downloads
• Multiple Datasets: Support for KITTI (more coming soon)
• Flexible Storage: Download to any directory
• Resume Support: Interrupted downloads can be resumed
• Minimal Dependencies: Lightweight with NumPy and Click

---

Installation

pip install mobility-datasets

---

Quick Start

Download via CLI

# Download KITTI GPS/IMU data and ground truth
mdb dataset download kitti --components oxts,poses

# Download complete KITTI dataset (~165 GB)
mdb dataset download kitti --all

# Download to custom directory
mdb dataset download kitti --components oxts --data-dir /mnt/datasets

Download via Python

from mobility_datasets.kitti.loader import KITTIDownloader

# Initialize downloader
downloader = KITTIDownloader(data_dir="./data/kitti")

# Download specific components
downloader.download(["oxts", "poses"], keep_zip=False)

# Or download everything
downloader.download_all(keep_zip=False)

---

Supported Datasets

Dataset	Status	Components
KITTI	✅ Available	oxts, poses, calib, sequences
nuScenes	🚧 Planned	-
Waymo	🚧 Planned	-

---

KITTI Dataset Components

Component	Description	Size	Frequency
oxts	GPS/IMU sensor data	~850 MB	10-100 Hz
poses	Ground truth trajectories	~5 MB	~10 Hz
calib	Calibration files	~10 MB	Static
sequences	Camera images, timestamps	~100 GB	~10 Hz

Total KITTI size: ~165 GB (all components)

---

Documentation

Full documentation available at: Read the Docs (coming soon)

• Quick Start: Get running in 5 minutes
• CLI Reference: Complete command-line guide
• API Reference: Python API documentation
• KITTI Guide: Dataset structure and usage

---

Requirements

• Python 3.11+
• NumPy
• Click
• Requests

---

Contributing

Contributions welcome! Please:

1. Fork the repository
2. Create a feature branch
3. Follow the Documentation Standards
4. Submit a pull request

---

License

MIT License - see LICENSE file for details

---

Citation

If you use this library in your research, please cite the original datasets:

KITTI Dataset:

@INPROCEEDINGS{Geiger2012CVPR,
  author = {Andreas Geiger and Philip Lenz and Raquel Urtasun},
  title = {Are we ready for Autonomous Driving? The KITTI Vision Benchmark Suite},
  booktitle = {Conference on Computer Vision and Pattern Recognition (CVPR)},
  year = {2012}
}

---

Acknowledgments

• KITTI Vision Benchmark Suite
• All dataset providers for making their data publicly available