Synset Signset Germany

Synset Signset Germany: A Synthetic Dataset for German Traffic Sign Recognition (Sielemann, Loercher et al., 2024)

The synthetic Synset Signset Germany dataset addresses the task of traffic sign recognition for the country of Germany. In this, it combines the advantages of data-driven and analytical modeling: GAN-based texture generation enables data-driven dirt and wear artifacts to create unique and realistic traffic sign surfaces, while the analytical scene modulation achieves physically correct lighting along with proper geometric transformations, and allows detailed parameterization.

The resulting synthetic traffic sign recognition dataset Synset Signset Germany contains a total of 105,500 images of 211 different German traffic sign classes, including newly issued (2020) and thus comparatively rare traffic signs. In addition to a mask and a segmentation image, we also provide extensive metadata, including the stochastically selected environment and imaging effect parameters for each image. Overall, the resulting dataset is among the largest and most diverse datasets for traffic sign recognition and, to the best of our knowledge, one of the first publicly available large-scale synthetic datasets for this task.

A subset of 43 classes in the dataset aims to represent a “synthetic twin” of the well-known “German Traffic Sign Recognition Benchmark“ (GTSRB)¹ dataset with similar imaging parameters. Overall, our dataset is therefore well suited for training traffic sign recognition applications or comparing real-world data with synthetic data. Thanks to the extensive metadata, it can also be used for applications in the context of explainable AI (XAI) or robustness analyses and systematic tests.

Details

Field	Task	License	Samples	Key Annotations	Institutions
Computer Vision	traffic sign recognition	Creative Commons Attribution 4.0	105,500 (independent images)	Per frame: Class (sign type), environment conditions (e.g., time of day, contrast), imaging artifacts (noise, motion blur, chromatic aberration), upper/lower sign, semantic segmentation, binary mask.	Fraunhofer IOSB, Fraunhofer IPA, KIT-IES

Paper and reference

This work is licensed under Creative Commons Attribution 4.0.

To cite this dataset in your scientific work, please use the following bibliography entry:

Sielemann, A., Loercher, L., Schumacher, M., Wolf, S., Roschani, M., Ziehn, J., and Beyerer, J. (2024). Synset Signset Germany: A Synthetic Dataset for German Traffic Sign Recognition. In 2024 IEEE 27th International Conference on Intelligent Transportation Systems (ITSC).

@inproceedings{synset_signset_ger_sielemann_2024,
  title={{Synset Signset Germany: A Synthetic Dataset for German Traffic Sign Recognition}},
  author={Sielemann, Anne and Loercher, Lena and Schumacher, Max-Lion and Wolf, Stefan and Roschani, Masoud and Ziehn, Jens and Beyerer, Juergen},
  booktitle={2024 IEEE 27th International Conference on Intelligent Transportation Systems (ITSC)},
  year={2024}
}

In case of copying and redistributing, or publishing an adapted version of our dataset, please provide the name of our dataset, the creator names, a copyright notice, a link to this website, a license notice with link to the license, and, if changes were made, a disclaimer notice, and a short description of the applied changes. For example, as follows:

This work is based on the Synset Signset Germany Dataset
by Anne Sielemann, Lena Loercher, Max-Lion Schumacher, Stefan Wolf, Jens Ziehn, Masoud Roschani, and Juergen Beyerer,
© 2024 Fraunhofer IOSB, All rights reserved.
Link: synset.de/datasets/synset-signset-ger/
Licence: CC BY 4.0
Disclaimer: The original authors are neither affiliated nor responsible for any applied changes.

Download Dataset    Download Paper for Personal Use

External platforms
Mobilithek entry (German)

Credits

This work was supported by the Fraunhofer Internal Programs under Grant No. PREPARE 40-02702 within the ML4Safety project, as well as funded by the German Federal Ministry for Economic Affairs and Climate Action (BMWK) within the program “New Vehicle and System Technologies” as part of the AVEAS research project. The real-world data of worn / dirty traffic signs, used to train the GAN for texture synthesis, was acquired with the kind support of the civil engineering department of the city of Karlsruhe, Germany (Tiefbauamt Karlsruhe).

1. J. Stallkamp, M. Schlipsing, J. Salmen, and C. Igel, “The German traffic sign recognition benchmark: a multi-class classification competition,” in The 2011 international joint conference on neural networks. IEEE, 2011, pp. 1453–1460. ↩︎
2. Verkehrszeichen Wissensnetz – stvo2Go ↩︎