• 輸入節(jié)點(diǎn)名字：“images”；數(shù)據(jù)：float32[1,3,640,640]。

• 輸出節(jié)點(diǎn)1的名字：“output0”；數(shù)據(jù)：float32[1,116,8400]。其中116的前84個(gè)字段跟 YOLOv8目標(biāo)檢測(cè)模型輸出定義完全一致，即cx,cy,w,h和80類的分?jǐn)?shù)；后32個(gè)字段用于計(jì)算掩膜數(shù)據(jù)。

• 輸出節(jié)點(diǎn)2的名字：“output1”；數(shù)據(jù)：float32[1,32,160,160]。output0后32個(gè)字段與output1的數(shù)據(jù)做矩陣乘法后得到的結(jié)果，即為對(duì)應(yīng)目標(biāo)的掩膜數(shù)據(jù)。

# Initialize the VideoCapture

cap =cv2.VideoCapture("store-aisle-detection.mp4")

# Initialize YOLOv5 Instance Segmentator

model_path ="yolov8n-seg.xml"

device_name ="GPU"

yoloseg =YOLOSeg(model_path, device_name, conf_thres=0.3, iou_thres=0.3)

whilecap.isOpened():

# Read frame from the video

ret, frame =cap.read()

ifnotret:

break

# Update object localizer

start =time.time()

boxes, scores, class_ids, masks =yoloseg(frame)

# postprocess and draw masks

combined_img =yoloseg.draw_masks(frame)

end =time.time()

# show FPS

fps =(1/(end -start))

fps_label ="Throughput: %.2fFPS"%fps

cv2.putText(combined_img, fps_label, (10, 25), cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 0, 255), 2)

# show ALL

cv2.imshow("YOLOv8 Segmentation OpenVINO inference Demo", combined_img)

# Press Any key stop

ifcv2.waitKey(1) >-1:

print("finished by user")

break