README.md

---
annotations_creators: []
language: en
license: cc-by-nc-sa-4.0
size_categories:
- n<1K
task_categories:
- object-detection
task_ids: []
pretty_name: VLM-GIST
tags:
- fiftyone
- image
- object-detection
description: This dataset contains images annotated with unique and semantically rich
  descriptions, as well as bounding boxes and segmentation masks for all visible object
  instances. Images '00013.jpg' to '00054.jpg' are taken from the AgiBot World dataset.
dataset_summary: '

  [*Leveraging Vision-Language Models for Open-Vocabulary Instance Segmentation and Tracking*](https://vlm-gist.github.io) by Bastian Pätzold, Jan Nogga and [Sven Behnke](https://www.ais.uni-bonn.de/behnke). IEEE Robotics and Automation Letters (RA-L). 2025.

  ![image/png](dataset_preview.png)



  This is a [FiftyOne](https://github.com/voxel51/fiftyone) dataset with 64 samples, thoroughly annotated with unique and semantically rich descriptions, as well as bounding boxes and segmentation masks for all visible object instances.


  ## 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("ptzld/VLM-GIST")


  # Launch the App

  session = fo.launch_app(dataset)

  ```

  '
---

# Dataset Card for VLM-GIST

Supplementary material of: [*Leveraging Vision-Language Models for Open-Vocabulary Instance Segmentation and Tracking*](https://vlm-gist.github.io) by Bastian Pätzold, Jan Nogga and [Sven Behnke](https://www.ais.uni-bonn.de/behnke). IEEE Robotics and Automation Letters (RA-L). 2025.

![image/png](dataset_preview.png)

This is a [FiftyOne](https://github.com/voxel51/fiftyone) dataset with 64 samples, thoroughly annotated with unique and semantically rich descriptions, as well as bounding boxes and segmentation masks for all visible object instances.

## 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("ptzld/VLM-GIST")

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

### Citation

If you utilize this dataset in your research, please cite:

https://arxiv.org/abs/2503.16538
```bibtex
@article{paetzold25vlmgist,
    author={Bastian P{\"a}tzold and Jan Nogga and Sven Behnke},
    title={Leveraging Vision-Language Models for Open-Vocabulary Instance Segmentation and Tracking},
    journal={IEEE Robotics and Automation Letters (RA-L)},
    volume={10},
    number={11},
    pages={11578-11585},
    year={2025}
}
```

### License

The VLM-GIST dataset is licensed under [CC BY‑NC‑SA 4.0](https://creativecommons.org/licenses/by-nc-sa/4.0/) (see [license](./LICENSE)), where 42 images ([00013.jpg](./data/00013.jpg) to [00054.jpg](./data/00054.jpg)) are taken from the [AgiBot World](https://agibot-world.com/) dataset.

### Contact

Bastian Pätzold <paetzold@ais.uni-bonn.de><br>
Jan Nogga <nogga@ais.uni-bonn.de>