Segformer B0 Person Segmentation
模型简介
模型特点
模型能力
使用案例
许可证:开放铁路 语言:
- 英语 任务标签:图像分割
描述
语义分割是一种计算机视觉技术,用于为图像中的每个像素分配一个标签,表示图像中物体或区域的语义类别。这是一种密集预测形式,因为它涉及为图像中的每个像素分配标签,而不仅仅是像目标检测或实例分割那样围绕物体或关键点的框。语义分割的目标是识别和理解图像中的物体和场景,并将图像分割成对应于不同实体的部分。
参数
model = SegformerForSemanticSegmentation.from_pretrained("/notebooks/segformer_5_epoch",
num_labels=2,
id2label=id2label,
label2id=label2id, )
使用方法
from torch import nn
import numpy as np
import matplotlib.pyplot as plt
# 转换
_transform = A.Compose([
A.Resize(height=512, width=512),
ToTensorV2(),
])
trans_image = _transform(image=np.array(image))
outputs = model(trans_image['image'].float().unsqueeze(0))
logits = outputs.logits.cpu()
print(logits.shape)
# 首先,将logits调整到原始图像大小
upsampled_logits = nn.functional.interpolate(logits,
size=image.size[::-1], # (高度, 宽度)
mode='bilinear',
align_corners=False)
seg = upsampled_logits.argmax(dim=1)[0]
color_seg = np.zeros((seg.shape[0], seg.shape[1], 3), dtype=np.uint8) # 高度, 宽度, 3
palette = np.array([[0, 0, 0],[255, 255, 255]])
for label, color in enumerate(palette):
color_seg[seg == label, :] = color
# 转换为BGR
color_seg = color_seg[..., ::-1]
<img src='data:image/png;base64,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











