#### NOTE
This is a **Hugging Face dataset**. For large datasets, ensure `huggingface_hub>=1.1.3` to avoid rate limits. Learn more in the <a href="https://docs.voxel51.com/integrations/huggingface.html#loading-datasets-from-the-hub" target="_blank">Hugging Face integration docs</a>.

<a href="https://huggingface.co/datasets/Voxel51/TAMPAR" target="_blank">![Hugging Face](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Dataset-yellow)</a>

# Dataset Card for TAMPAR

![image/png](https://huggingface.co/datasets/Voxel51/TAMPAR/resolve/main/tampar-skeletons.gif)

This is a [FiftyOne](https://github.com/voxel51/fiftyone) dataset with 485 samples.

The samples here are from the test set.

## Installation

If you haven’t already, install FiftyOne:

```bash
pip install -U fiftyone
```

## Usage

```python
import fiftyone as fo
from fiftyone.utils.huggingface import load_from_hub

# Load the dataset
# Note: other available arguments include 'max_samples', etc
dataset = load_from_hub("voxel51/TAMPAR")

# Launch the App
session = fo.launch_app(dataset)
```

## Dataset Details

### Dataset Description

TAMPAR is a novel real-world dataset of parcels

- with >900 annotated real-world images with >2,700 visible parcel side surfaces,
- 6 different tampering types, and
- 6 different distortion strengths

This dataset was collected as part of the WACV ‘24 [paper](https://arxiv.org/abs/2311.03124)  *“TAMPAR: Visual Tampering Detection for Parcels Logistics in Postal Supply Chains”*

- **Curated by:** Alexander Naumann, Felix Hertlein, Laura Dörr and Kai Furmans
- **Funded by:** FZI Research Center for Information Technology, Karlsruhe, Germany
- **Shared by:** [Harpreet Sahota](https://huggingface.co/harpreetsahota), Hacker-in-Residence at Voxel51
- **License:** [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/legalcode)

### Dataset Sources

- **Repository:** https://github.com/a-nau/tampar
- **Paper:** https://arxiv.org/abs/2311.03124
- **Demo:** https://a-nau.github.io/tampar/

## Uses

### Direct Use

Multisensory setups within logistics facilities and a simple cell phone camera during the last-mile delivery, where only a single RGB image is taken and compared against a reference from an existing database to detect potential appearance changes that indicate tampering.

## Dataset Structure

COCO Format Annotations

## Citation

```bibtex
@inproceedings{naumannTAMPAR2024,
    author    = {Naumann, Alexander and Hertlein, Felix and D\"orr, Laura and Furmans, Kai},
    title     = {TAMPAR: Visual Tampering Detection for Parcels Logistics in Postal Supply Chains},
    booktitle = {Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision},
    month     = {January},
    year      = {2024},
    note      = {to appear in}
}
```