AksharPatel/manga-speech-bubble-segmentation

This model uses YOLO v8 to train the model.

from ultralytics import YOLO
import cv2

model = YOLO("model/best.pt")
image = cv2.imread("test.jpg")
h, w, c = image.shape

results = model(
    image,
    # dimensions required to be multiple of 32
    imgsz=[(h // 32) * 32, (w // 32) * 32],
    conf=0.6,
)[0]
# save the image with bounding boxes
results.save("output.jpg")