🚀 Segformer模型卡
Segformer是用于图像分割的模型,通过预训练模型可快速进行图像分割推理,支持在不同设备上运行。
🚀 快速开始
加载预训练模型
点击下面的按钮在Colab中打开示例:

步骤如下
- 安装所需依赖:
pip install -U segmentation_models_pytorch albumentations
- 运行推理代码:
import torch
import requests
import numpy as np
import albumentations as A
import segmentation_models_pytorch as smp
from PIL import Image
device = "cuda" if torch.cuda.is_available() else "cpu"
checkpoint = "smp-hub/segformer-b5-1024x1024-city-160k"
model = smp.from_pretrained(checkpoint).eval().to(device)
preprocessing = A.Compose.from_pretrained(checkpoint)
url = "https://huggingface.co/datasets/hf-internal-testing/fixtures_ade20k/resolve/main/ADE_val_00000001.jpg"
image = Image.open(requests.get(url, stream=True).raw)
np_image = np.array(image)
normalized_image = preprocessing(image=np_image)["image"]
input_tensor = torch.as_tensor(normalized_image)
input_tensor = input_tensor.permute(2, 0, 1).unsqueeze(0)
input_tensor = input_tensor.to(device)
with torch.no_grad():
output_mask = model(input_tensor)
mask = torch.nn.functional.interpolate(
output_mask, size=(image.height, image.width), mode="bilinear", align_corners=False
)
mask = mask.argmax(1).cpu().numpy()
💻 使用示例
基础用法
上述“加载预训练模型”中的代码即为基础的推理使用示例。
高级用法
暂未提供高级用法示例。
📚 详细文档
模型初始化参数
model_init_params = {
"encoder_name": "mit_b5",
"encoder_depth": 5,
"encoder_weights": None,
"decoder_segmentation_channels": 768,
"in_channels": 3,
"classes": 19,
"activation": None,
"aux_params": None
}
数据集
数据集名称:Cityscapes
更多信息
- 库:https://github.com/qubvel/segmentation_models.pytorch
- 文档:https://smp.readthedocs.io/en/latest/
- 许可证:https://github.com/NVlabs/SegFormer/blob/master/LICENSE
此模型已使用 PytorchModelHubMixin 推送到模型中心。
📄 许可证
本项目使用其他许可证,详情请见:https://github.com/NVlabs/SegFormer/blob/master/LICENSE