Mediated Machine Vision

Making it easier to build taxon identification models using GBIF-mediated data for use in artificial intelligence applications

Model

Danish Mycological Society, Google Research. Danish Mycological Society Fungi Embedding Model. TensorFlow Hub. https://doi.org/10.26161/dhpb-3346

License: CC BY-NC 4.0

This model is unsuitable for:

Distinguishing edible and inedible mushrooms.
Using in foraging applications.

Data source

Frøslev T G, Heilmann-Clausen J, Lange C, Læssøe T, Petersen J H, Søchting U, Jeppesen T S, Vesterholt J (2019). Danish Mycological Society, fungal records database. Danish Mycological Society. Occurrence dataset https://doi.org/10.15468/zn159h accessed via GBIF.org on 2019-10-16.

Source data: Danish Mycological Society, fungal records database

License: CC BY-NC 4.0

Limitations

This model may not generalize to cultivated or harvested mushrooms.
This model may not generalize to mushroom species common to other regions.
This model may not generalize to fungi species in non-user-captured images, such as museum specimens.
This model may not generalize to natural world species outside of the fungi kingdom.