--- tags: - image-classification - ecology - fungi - FGVC library_name: FungiTastic Dataset license: cc-by-nc-4.0 --- # Model card for BVRA/tf_efficientnet_b3.in1k_ft_fungitastic-mini_224 ## Model Details - **Model Type:** Fine-grained classification of fungi species - **Model Stats:** - Params (M): 11.0 - Image size: 224 x 224 - **Papers:** - **Original:** --> ??? - **Train Dataset:** FungiTastic --> https://arxiv.org/pdf/2408.13632 ## Model Usage ### Image Embeddings ```python import timm import torch import torchvision.transforms as T from PIL import Image from urllib.request import urlopen model = timm.create_model("hf-hub:BVRA/tf_efficientnet_b3.in1k_ft_fungitastic-mini_224", pretrained=True) model = model.eval() train_transforms = T.Compose([T.Resize((224, 224)), T.ToTensor(), T.Normalize([0.5, 0.5, 0.5], [0.5, 0.5, 0.5])]) img = Image.open(PATH_TO_YOUR_IMAGE) output = model(train_transforms(img).unsqueeze(0)) # output is a (1, num_features) shaped tensor ``` ## Citation ```bibtex @article{picek2024fungitastic, title={FungiTastic: A multi-modal dataset and benchmark for image categorization}, author={Picek, Lukas and Janouskova, Klara and Sulc, Milan and Matas, Jiri}, journal={arXiv preprint arXiv:2408.13632}, year={2024} } ``` ```bibtex @InProceedings{Picek_2022_WACV, author = {Picek, Luk'a {s} and {S}ulc, Milan and Matas, Ji {r}{'\i} and Jeppesen, Thomas S. and Heilmann-Clausen, Jacob and L{e}ss{\o}e, Thomas and Fr{\o}slev, Tobias}, title = {Danish Fungi 2020 - Not Just Another Image Recognition Dataset}, booktitle = {Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)}, month = {January}, year = {2022}, pages = {1525-1535} } ``` ```bibtex @article{picek2022automatic, title={Automatic Fungi Recognition: Deep Learning Meets Mycology}, author={Picek, Luk{'a}{ {s}} and { {S}}ulc, Milan and Matas, Ji{ {r}}{'\i} and Heilmann-Clausen, Jacob and Jeppesen, Thomas S and Lind, Emil}, journal={Sensors}, volume={22}, number={2}, pages={633}, year={2022}, publisher={Multidisciplinary Digital Publishing Institute} } ```