Megadescriptor L 384

🚀 MegaDescriptor-L-384模型卡片

MegaDescriptor-L-384是一个基于Swin-L架构的图像特征模型，它在动物重识别数据集上进行了超智能的预训练，能够为动物图像的特征提取和重识别任务提供强大支持。

🚀 快速开始

MegaDescriptor-L-384是一个Swin-L图像特征模型，在动物重识别数据集上进行了预训练。下面是使用该模型生成图像嵌入的示例代码：

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/MegaDescriptor-L-384", pretrained=True)
model = model.eval()

train_transforms = T.Compose([T.Resize(size=(384, 384)),
                              T.ToTensor(), 
                              T.Normalize([0.5, 0.5, 0.5], [0.5, 0.5, 0.5])]) 

img = Image.open(urlopen(
    'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))

output = model(train_transforms(img).unsqueeze(0))  # output is (batch_size, num_features) shaped tensor
# output is a (1, num_features) shaped tensor

✨ 主要特性

• 适用领域广泛：适用于图像分类、生态学、动物识别和重识别等多个领域。
• 预训练优势：在动物重识别数据集上进行预训练，能更好地处理动物相关图像。

📚 详细文档

模型详情

模型使用

图像嵌入

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/MegaDescriptor-L-384", pretrained=True)
model = model.eval()

train_transforms = T.Compose([T.Resize(size=(384, 384)),
                              T.ToTensor(), 
                              T.Normalize([0.5, 0.5, 0.5], [0.5, 0.5, 0.5])]) 

img = Image.open(urlopen(
    'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))

output = model(train_transforms(img).unsqueeze(0))  # output is (batch_size, num_features) shaped tensor
# output is a (1, num_features) shaped tensor

引用信息

@inproceedings{vcermak2024wildlifedatasets,
  title={WildlifeDatasets: An open-source toolkit for animal re-identification},
  author={{\v{C}}erm{\'a}k, Vojt{\v{e}}ch and Picek, Lukas and Adam, Luk{\'a}{\v{s}} and Papafitsoros, Kostas},
  booktitle={Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision},
  pages={5953--5963},
  year={2024}
}

📄 许可证

本项目采用CC BY-NC 4.0许可证。