iciness/TruckV2X

---
license: mit
size_categories:
- 10K<n<100K
source_datasets:
- original
task_categories:
- object-detection
pretty_name: TruckV2X
tags:
- Cooperative Perception
- Autonomous Trucking
- Dataset
- V2X
library_name: datasets
---

# TruckV2X: A Truck-Centered Perception Dataset

[Paper](https://ieeexplore.ieee.org/abstract/document/11096563)

[Project Page](https://xietenghu1.github.io/TruckV2X/)

TruckV2X is the first large-scale truck-centered cooperative perception dataset, designed to address unique perception challenges in autonomous trucking (e.g., extensive blind spots, occlusions from large vehicle sizes, and dynamic trailer movements). It features multi-modal sensing (LiDAR and cameras) and supports multi-agent cooperation, including interactions between tractors, trailers, connected and automated vehicles (CAVs), and road-side units (RSUs). This dataset establishes performance benchmarks for heavy-duty vehicle scenarios and accelerates research on multi-agent autonomous trucking systems.

## Usage Instructions

### Option 1: Load via Hugging Face Hub

You can directly load the dataset using the `datasets` library for quick experimentation or exploration:

```python
from datasets import load_dataset

dataset = load_dataset("XieTenghu1/TruckV2X", trust_remote_code=True)
# For security reasons, 🤗 Datasets do not allow running dataset loading scripts by default,
# and you have to pass trust_remote_code=True to load datasets that require running a dataset script.
```

> **Note:** This method may be more user-friendly and convenient for quick access, but it does **not expose the raw file structure**, which is required by downstream frameworks like OpenCOOD for direct usage.

### Option 2: Clone + Local Extraction (Recommended)

For full flexibility and seamless integration with cooperative perception frameworks (e.g., OpenCOOD), we strongly recommend downloading and using the dataset locally:

```bash
# Clone the dataset repository
git lfs install
git clone https://huggingface.co/datasets/XieTenghu1/TruckV2X
cd TruckV2X

# Unzip all splits (each Town is separately zipped)
# Unzip all zip files under train/
for zipfile in train/*.zip; do
dirname="${zipfile%.zip}"
unzip "$zipfile" -d "$dirname"
done

# Unzip all zip files under val/
for zipfile in val/*.zip; do
dirname="${zipfile%.zip}"
unzip "$zipfile" -d "$dirname"
done

# Unzip all zip files under test/
for zipfile in test/*.zip; do
dirname="${zipfile%.zip}"
unzip "$zipfile" -d "$dirname"
done
```

## Dataset Structure

The dataset is organized into training, validation and testing subsets with the following structure:

```
TruckV2X/
├── train/
│ ├── Town1_1/
│ │ ├── cav/
│ │ │ ├── 000000_camera0.jpg
│ │ │ ├── 000000_camera1.jpg
│ │ │ ├── 000000_camera2.jpg
│ │ │ ├── 000000_camera3.jpg
│ │ │ ├── 000000_lidar0.pcd
│ │ │ ├── 000000.yaml
│ │ │ ├── 000001_xxx
│ │ │ ├── ...
│ │ ├── roadside/
│ │ │ ├── 000000_camera0.jpg
│ │ │ ├── 000000_camera1.jpg
│ │ │ ├── 000000.yaml
│ │ │ ├── 000001_xxx
│ │ │ ├── ...
│ │ ├── tractor/
│ │ │ ├── 000000_camera0.jpg
│ │ │ ├── 000000_camera1.jpg
│ │ │ ├── 000000_camera2.jpg
│ │ │ ├── 000000_camera3.jpg
│ │ │ ├── 000000_camera4.jpg
│ │ │ ├── 000000_lidar0.pcd
│ │ │ ├── 000000_lidar1.pcd
│ │ │ ├── 000000.yaml
│ │ │ ├── 000001_xxx
│ │ │ ├── ...
│ │ ├── trailer/
│ │ │ ├── 000000_camera0.jpg
│ │ │ ├── 000000_camera1.jpg
│ │ │ ├── 000000_camera2.jpg
│ │ │ ├── 000000_camera3.jpg
│ │ │ ├── 000000_camera4.jpg
│ │ │ ├── 000000_lidar0.pcd
│ │ │ ├── 000000_lidar1.pcd
│ │ │ ├── 000000.yaml
│ │ │ ├── 000001_xxx
│ │ │ ├── ...
│ ├── Town1_4/
│ ├── ...
├── val/
│ ├── Town3_0/
│ ├── ...
├── test/
│ ├── Town1_0/
│ ├── ...
```

## License

This dataset is licensed under the [MIT License](https://opensource.org/licenses/MIT).

## Citation

If you use TruckV2X or find our work inspiring in your research, please cite our paper:
```
@ARTICLE{11096563,
author={Xie, Tenghui and Song, Zhiying and Wen, Fuxi and Li, Jun and Liu, Guangzhao and Zhao, Zijian},
journal={IEEE Robotics and Automation Letters},
title={TruckV2X: A Truck-Centered Perception Dataset},
year={2025},
volume={10},
number={9},
pages={9312-9319},
keywords={Agricultural machinery;Laser radar;Cameras;Robot sensing systems;Roads;Vehicle dynamics;Sensors;Safety;Vehicle-to-infrastructure;Training;Vehicle-to-everything;cooperative perception;autonomous trucking;dataset},
doi={10.1109/LRA.2025.3592884}}
```