最近需要实例分割完成一些任务，一直用的SAM(segment anything）速度慢，找一个轻量分割模型。

1. YOLO-v8-seg使用

git clone https://github.com/ultralytics/ultralytics.git

cd ultralytics

vim run.py

from ultralytics import YOLO

# Load a model
model = YOLO('yolov8l-seg.pt')  # load an official model

# Predict with the model
results = model('test.jpg')  # predict on an image

就这么几行，这就行了？代码量也太少了。

2 效果

效果也还不错

速度：

3. 对结果results的解析

可以看官方详细介绍：预测 -Ultralytics YOLOv8 文档

img = cv2.imread(img_path)
imgray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)
print(img.shape)
for result in results:
    boxes = result.boxes  # Boxes object for bounding box outputs
    masks = result.masks  # Masks object for segmentation masks outputs
    for mask in masks:
        print(mask.xy)
        print(type(mask.xy))
        points = np.array(mask.xy, dtype=np.int32)
        cv2.polylines(img, [points], isClosed=True, color=(0, 255, 0), thickness=2)
    keypoints = result.keypoints  # Keypoints object for pose outputs
    probs = result.probs  # Probs object for classification outputs
    #result.show()  # display to screen
    result.save(filename='result.jpg')  # save to disk
cv2.imwrite('test.jpg',img)

很好用，希望对你有帮助！

THE END！