PhysicalAI-Robotics-GR00T-Eval

<div style="display: flex;">
<img src="https://huggingface.co/datasets/nvidia/EVAL-175/resolve/main/gr1_behavior/0_Open%20the%20box.png" width="20%"/>
<img src="https://huggingface.co/datasets/nvidia/EVAL-175/resolve/main/gr1_env/14_Use%20the%20right%20hand%20to%20pick%20up%20paper%20cup%20to%20metal%20container.png" width="20%">
<img src="https://huggingface.co/datasets/nvidia/EVAL-175/resolve/main/gr1_env/28_Use%20the%20right%20hand%20to%20pick%20up%20the%20wooden%20spoon%20and%20perform%20a%20scooping%20motion%20between%20the%20two%20bowls.png" width="20%"/>
<img src="https://huggingface.co/datasets/nvidia/EVAL-175/resolve/main/gr1_env/3_Use%20the%20right%20hand%20to%20close%20the%20lid%20of%20the%20soup%20tureen.png" width="20%"/>
<img src="https://huggingface.co/datasets/nvidia/EVAL-175/resolve/main/gr1_object/0_Use%20the%20left%20hand%20to%20pick%20up%20dark%20green%20cucumber%20from%20on%20circular%20gray%20mat%20to%20above%20beige%20bowl..png" width="20%"/>
</div>

# EVAL-175

## Dataset Description:
123 initial frame pictures from the robot's perspective before performing various tasks in the lab.

This dataset is ready for commercial/non-commercial use.

## Dataset Owner(s):
NVIDIA Corporation (GEAR Lab)

## Dataset Creation Date:
May 1, 2025

## License/Terms of Use:
This dataset is governed by the Creative Commons Attribution 4.0 International License (CC-BY-4.0).<br>
This dataset was created using a Fourier GR1-T2 robot.

## Intended Usage:
Training image dataset for robotics.

## Dataset Characterization
**Data Collection Method**<br>
- Human <br>

**Labeling Method**<br>
- Human <br>

## Dataset Format
- PNG files<br>
- TXT files

## Dataset Quantification

Record Count - 123 images
181.1MB

## Reference(s):

[DreamGen WhitePaper](https://arxiv.org/html/2505.12705v1)<br>
[IDM-GR1 Model](https://huggingface.co/seonghyeonye/IDM_gr1)

## Ethical Considerations:
NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. When downloaded or used in accordance with our terms of service, developers should work with their internal model team to ensure this model meets requirements for the relevant industry and use case and addresses unforeseen product misuse.

Please report security vulnerabilities or NVIDIA AI Concerns [here](https://www.nvidia.com/en-us/support/submit-security-vulnerability/).

<div style="display: flex;">
<img src="https://huggingface.co/datasets/nvidia/EVAL-175/resolve/main/gr1_behavior/0_Open%20the%20box.png" width="20%" alt="打开箱子"/>
<img src="https://huggingface.co/datasets/nvidia/EVAL-175/resolve/main/gr1_env/14_Use%20the%20right%20hand%20to%20pick%20up%20paper%20cup%20to%20metal%20container.png" width="20%" alt="使用右手将纸杯拿起并放置于金属容器中"/>
<img src="https://huggingface.co/datasets/nvidia/EVAL-175/resolve/main/gr1_env/28_Use%20the%20right%20hand%20to%20pick%20up%20the%20wooden%20spoon%20and%20perform%20a%20scooping%20motion%20between%20the%20two%20bowls.png" width="20%" alt="使用右手拿起木勺并在两个碗之间完成舀取动作"/>
<img src="https://huggingface.co/datasets/nvidia/EVAL-175/resolve/main/gr1_env/3_Use%20the%20right%20hand%20to%20close%20the%20lid%20of%20the%20soup%20tureen.png" width="20%" alt="使用右手合上汤盆的盖子"/>
<img src="https://huggingface.co/datasets/nvidia/EVAL-175/resolve/main/gr1_object/0_Use%20the%20left%20hand%20to%20pick%20up%20dark%20green%20cucumber%20from%20on%20circular%20gray%20mat%20to%20above%20beige%20bowl..png" width="20%" alt="使用左手从圆形灰色垫子上拿起深绿色黄瓜并放置于米色碗上方"/>
</div>

# EVAL-175

## 数据集描述
本数据集包含123张以机器人视角拍摄的实验室各类任务执行前的初始帧图像。
本数据集可用于商业及非商业用途。

## 数据集所有者
英伟达公司（GEAR实验室）

## 数据集创建日期
2025年5月1日

## 许可/使用条款
本数据集遵循知识共享署名4.0国际许可协议（Creative Commons Attribution 4.0 International License，CC-BY-4.0）进行管理。本数据集基于傅里叶GR1-T2机器人构建。

## 预期用途
面向机器人领域的训练用图像数据集。

## 数据集特征
### 数据采集方法
- 人工采集
### 标注方法
- 人工标注

## 数据集格式
- PNG图像文件
- TXT文本文件

## 数据集规模
图像总数：123张
总文件大小：181.1MB

## 参考文献
[DreamGen 白皮书](https://arxiv.org/html/2505.12705v1)
[IDM-GR1 模型](https://huggingface.co/seonghyeonye/IDM_gr1)

## 伦理考量
英伟达认为，可信人工智能是一项共同责任，我们已建立相关政策与实践，以支持各类人工智能应用的开发。开发者在按照本服务条款下载或使用本数据集时，应与其内部模型团队协作，确保该数据集符合相关行业与应用场景的要求，并防范可能出现的产品误用问题。

如需报告安全漏洞或英伟达人工智能相关问题，请[点击此处](https://www.nvidia.com/en-us/support/submit-security-vulnerability/)提交。