Search

Start Over Contributor Czech Ministry of Education, Youth and Sports@@LO1506@@Sustainability support of the centre NTIS - New Technologies for the Information Society@@nationalFunds@@ Contributor Czech Technical University in Prague@@SGS20/171/OHK3/3T/13@@Studentská Grantová Soutež@@Other@@ Type IMAGE Original context has metadata only false

1. Danish Fungi 2020

Creator:
Picek, Lukáš, Šulc, Milan, Matas, Jiří, Jeppesen, Thomas S., Heilmann-Clausen, Jacob, Læssøe, Thomas, and Frøslev, Tobias
Publisher:
IEEE/CVF
Type:
IMAGE and corpus
Subject:
Fungi, image processing, Classification, and Fine-grained
Description:
Danish Fungi 2020 (DF20) is a fine-grained dataset and benchmark. The dataset, constructed from observations submitted to the Danish Fungal Atlas, is unique in its taxonomy-accurate class labels, small number of errors, highly unbalanced long-tailed class distribution, rich observation metadata, and well-defined class hierarchy. DF20 has zero overlap with ImageNet, allowing unbiased comparison of models fine-tuned from publicly available ImageNet checkpoints. The dataset has 1,604 different classes, with 248,466 training images and 27,608 test images.
Rights:
GNU Library or "Lesser" General Public License 3.0 (LGPL-3.0), http://opensource.org/licenses/LGPL-3.0, and PUB