HOWS-CL-25

HOWS-CL-25 is a synthetic dataset especially designed for object classification on mobile robots operating in a changing environment (like a household), where it is important to learn new, never seen objects on the fly.

All 25 categories used in the HOWS-CL-25 dataset
DownloadDownload

Details

HOWS-CL-25 (Household Objects Within Simulation dataset for Continual Learning) is a synthetic dataset especially designed for object classification on mobile robots operating in a changing environment (like a household), where it is important to learn new, never seen objects on the fly. This dataset can also be used for other learning use-cases, like instance segmentation or depth estimation. Or where household objects or continual learning are of interest. HOWS contains 150795 RGB images containing 25 categories, over 925 instances of household objects, and corresponding normal, depth, and segmentation images. The dataset was created using Blenderproc and can be downloaded from Zenodo.

For further information, please have a look on our GitHub-Repository and our IROS-Paper.

Download link: https://zenodo.org/record/7189434

License (for files): https://creativecommons.org/licenses/by/4.0/legalcode

If you use HOWS in a research project or publication, please cite as follows:

@article{knauer2022recall, title={RECALL: Rehearsal-free Continual Learning for Object Classification}, author={Knauer, Markus and Denninger, Maximilian and Triebel, Rudolph}, journal={IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)}, doi={10.1109/IROS47612.2022.9981968} year={2022} }

@dataset{knauer2022hows, title={HOWS-CL-25: Household Objects Within Simulation Dataset for Continual Learning}, author={Knauer, Markus and Denninger, Maximilian and Triebel, Rudolph}, publisher={Zenodo}, year={2022} doi={10.5281/zenodo.7189434} url={https://doi.org/10.5281/zenodo.7189434 }

Publications

  • https://github.com/DLR-RM/RECALL
  • Markus Wendelin Knauer, Maximilian Denninger, Rudolph Triebel, "RECALL: Rehearsal-free Continual Learning for Object Classification", in: IEEE/RSJ. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2022), 24-26 Oct 2022, Kyoto, Japan [elib]
  • Maximilian Denninger, Dominik Winkelbauer, Martin Sundermeyer, Wout Boerdijk, Markus Knauer, Markus Wendelin, Klaus H. Strobl, Matthias Humt, Rudolph Triebel,"BlenderProc2: A Procedural Pipeline for Photorealistic Rendering", 8 (82), p. 4901, Februar, 2023. [elib]
  • Markus Wendelin Knauer, Maximilian Denninger, Rudolph Triebel, "RECALL: Rehearsal-free Continual Learning for Object Classification", in: IEEE/RSJ. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2022), 24-26 Oct 2022, Kyoto, Japan [elib]