Important: this repository does not contain the actual PInSoRo data
(~2TB would not work with GitHub!), only the meta data. See the PInSoRo web
site for details to access the actual
data.
This dataset contains 75 recordings of free-play interactions between 2 children (45 recordings) or one child and one robot (30 recordings).
To learn more about the dataset, visit its web site.
You are very welcome to use this dataset for your own research. If you do so, please cite the The PInSoRo dataset: Supporting the data-driven study of child-child and child-robot social dynamics article.
- Download and unzip the lastest version of the dataset from the Zenodo platform
- Launch
python ./tools/visualise_dataset.py ./data/<record>/pinsoro-*.csv. If everything is working as expect, this should replay the interaction as seen in the screenshot above. - Have fun! (and do not hesitate to use the source of
visualise_dataset.pyas a starting point for your own research)
More in depth, this repository contains two things:
- the dataset metadata (and data once you have obtained the actual data)
in the
data/subdirectory. Check the README.md file there to know more about how the data is stored. - a few tools to explore and manipulate the dataset. Check the README.md to know more.
Advanced analysis tools are available as well in a separate repository. Check the dataset's website to know more about those.
If you have specific questions about the dataset, you can contact Séverin Lemaignan. If your question might be relevant to other people, please instead open an issue.