TY - GEN

T1 - Danish Fungi 2020 - Not Just Another Image Recognition Dataset

AU - Picek, Lukas

AU - Sulc, Milan

AU - Matas, Jiri

AU - Jeppesen, Thomas S.

AU - Heilmann-Clausen, Jacob

AU - Læssøe, Thomas

AU - Frøslev, Tobias

N1 - Publisher Copyright: © 2022 IEEE.

PY - 2022

Y1 - 2022

N2 - We introduce a novel fine-grained dataset and bench-mark, the Danish Fungi 2020 (DF20). The dataset, constructed from observations submitted to the Atlas of Danish Fungi, 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, al-lowing unbiased comparison of models fine-tuned from publicly available ImageNet checkpoints. The proposed evaluation protocol enables testing the ability to improve classification using metadata - e.g. precise geographic location, habitat, and substrate, facilitates classifier calibration testing, and finally allows to study the impact of the device settings on the classification performance. Experiments using Convolutional Neural Networks (CNN) and the recent Vision Transformers (ViT) show that DF20 presents a challenging task. Interestingly, ViT achieves results su-perior to CNN baselines with 80.45% accuracy and 0.743 macro F1 score, reducing the CNN error by 9% and 12% respectively. A simple procedure for including metadata into the decision process improves the classification accuracy by more than 2.95 percentage points, reducing the error rate by 15%. The source code for all methods and experiments is available at https://sites.google.com/view/danish-fungi-dataset.

AB - We introduce a novel fine-grained dataset and bench-mark, the Danish Fungi 2020 (DF20). The dataset, constructed from observations submitted to the Atlas of Danish Fungi, 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, al-lowing unbiased comparison of models fine-tuned from publicly available ImageNet checkpoints. The proposed evaluation protocol enables testing the ability to improve classification using metadata - e.g. precise geographic location, habitat, and substrate, facilitates classifier calibration testing, and finally allows to study the impact of the device settings on the classification performance. Experiments using Convolutional Neural Networks (CNN) and the recent Vision Transformers (ViT) show that DF20 presents a challenging task. Interestingly, ViT achieves results su-perior to CNN baselines with 80.45% accuracy and 0.743 macro F1 score, reducing the CNN error by 9% and 12% respectively. A simple procedure for including metadata into the decision process improves the classification accuracy by more than 2.95 percentage points, reducing the error rate by 15%. The source code for all methods and experiments is available at https://sites.google.com/view/danish-fungi-dataset.

KW - Datasets

KW - Evaluation and Comparison of Vision Algorithms Object Detection/Recognition/Categorization

U2 - 10.1109/WACV51458.2022.00334

DO - 10.1109/WACV51458.2022.00334

M3 - Article in proceedings

AN - SCOPUS:85122859027

SP - 3281

EP - 3291

BT - Proceedings - 2022 IEEE Winter Conference on Applications of Computer Vision, WACV 2022

A2 - O'Conner, Lisa

PB - IEEE

T2 - 22nd IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2022

Y2 - 4 January 2022 through 8 January 2022

ER -