ARMBENCH DATASET

News and updates

March, 2023: Benchmark datasets publicly released for 3 key perception challenges in robotic manipulation: (a) Object segmentation, (b) Object identification, and (c) Defect detection.

January, 2023: The ARMBench research paper has been accepted for presentation at ICRA, 2023.

What is ARMBench?

ARMBench is a large-scale benchmark dataset for perception and manipulation challenges in a robotic pick-and-place setting. The dataset is collected in an Amazon warehouse and captures a wide variety of objects and configurations. It comprises images and videos for different stages of robotic manipulation including picking, transferring, and placing with high-quality annotations.

Currently, the dataset provides data annotations for three main computer vision tasks: Object segmentation, Object Identification, and Defect Detection on images and videos.

Research Paper

Supplementary Video

Credits

This dataset is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0). Use the following citation if you use this dataset for publication.

@article{mitash2023armbench,
    title={ARMBench: An object-centric benchmark dataset for robotic manipulation},
    author={Mitash, Chaitanya and Wang, Fan and Lu, Shiyang and Terhuja, Vikedo,
            and Garaas, Tyler and Polido, Felipe and Nambi, Manikantan},
    journal={arXiv preprint arXiv:2303.16382},
    year={2023}
}