Yolov8 Human Detection Thermal

🚀 热成像相机人体检测

本项目提供了一个基于热成像图像的人体检测模型，可用于伪彩色和灰度热成像图像。模型最初针对人体检测进行了微调，也可进一步微调以检测其他物体。

🚀 快速开始

部署模型

要部署此模型，请使用以下代码：

安装依赖

$ python -m pip install ultralytics supervision huggingface_hub 

Python代码

# import libraries
from huggingface_hub import hf_hub_download
from ultralytics import YOLO
from supervision import Detections
import cv

# download model
model_path = hf_hub_download(
    repo_id = "pitangent-ds/YOLOv8-human-detection-thermal",
    filename = "model.pt"
)

# load model
model = YOLO(model_path)

# method for inference
def inference(image_path):
    cv_image = cv.imread(image_path, cv2.IMREAD_ANYCOLOR)
    model_output = model(cv_image, conf=0.6, verbose=False)
    detections = Detections.from_ultralytics(model_output[0])
    return detections

✨ 主要特性

• 多图像类型支持：支持伪彩色和灰度热成像图像。
• 可扩展性：模型最初针对人体检测进行微调，也可进一步微调以检测其他物体。

📦 安装指南

安装所需的依赖库：

$ python -m pip install ultralytics supervision huggingface_hub 

💻 使用示例

基础用法

# import libraries
from huggingface_hub import hf_hub_download
from ultralytics import YOLO
from supervision import Detections
import cv

# download model
model_path = hf_hub_download(
    repo_id = "pitangent-ds/YOLOv8-human-detection-thermal",
    filename = "model.pt"
)

# load model
model = YOLO(model_path)

# method for inference
def inference(image_path):
    cv_image = cv.imread(image_path, cv2.IMREAD_ANYCOLOR)
    model_output = model(cv_image, conf=0.6, verbose=False)
    detections = Detections.from_ultralytics(model_output[0])
    return detections

📚 详细文档

训练代码

• 数据集链接：Roboflow Universe

from ultralytics import YOLO
import torch

# load model
model = YOLO("yolov8n.pt")

# hyper parameters
hyperparams = {
    "batch": 32,
    "epochs": 30,
    "imgsz": [640, 480],
    "optimizer": "AdamW",
    "cos_lr": True,
    "lr0": 3e-5,
    "warmup_epochs": 10
}

# start training
model.train(
    device = 'cuda' if torch.cuda.is_available() else 'cpu',
    data = "data.yaml",
    **hyperparams
)

• 点击此处查看训练参数：Training Arguments

依赖库

python: 3.10.13
ultralytics: 8.0.206
torch: "2.1.0+cu118"
roboflow: 1.1.9

📄 许可证

本项目采用AGPL-3.0许可证。