File size: 1,366 Bytes
e811d70 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 | import sys
import cv2
import numpy as np
import onnxruntime as ort
MODEL_PATH = "cat_landmark_model.onnx"
def preprocess(img_path):
img = cv2.imread(img_path)
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
orig_h, orig_w = img.shape[:2]
img_resized = cv2.resize(img, (224, 224))
tensor = img_resized.astype(np.float32).transpose(2, 0, 1) / 255.0
return np.expand_dims(tensor, 0), orig_w, orig_h
def run(img_path):
session = ort.InferenceSession(MODEL_PATH)
tensor, orig_w, orig_h = preprocess(img_path)
outputs = session.run(None, {"input": tensor})[0][0] # shape: (18,)
landmarks = outputs.reshape(9, 2)
# denormalize back to original image coords
landmarks[:, 0] *= orig_w
landmarks[:, 1] *= orig_h
for i, (x, y) in enumerate(landmarks):
print(f"Point {i}: ({x:.1f}, {y:.1f})")
# draw bounding box around landmarks
img = cv2.imread(img_path)
x1, y1 = landmarks.min(axis=0).astype(int)
x2, y2 = landmarks.max(axis=0).astype(int)
cv2.rectangle(img, (x1, y1), (x2, y2), (0, 255, 0), 2)
for (x, y) in landmarks.astype(int):
cv2.circle(img, (x, y), 3, (0, 0, 255), -1)
out_path = "output.jpg"
cv2.imwrite(out_path, img)
print(f"Saved to {out_path}")
if __name__ == "__main__":
img_path = sys.argv[1] if len(sys.argv) > 1 else "test.jpg"
run(img_path)
|