Realised my mistake was using np.expand_dims instead of cv2.dnn.blobFromImage, here is the working code for anyone with a similar issue:
```
import cv2
from openvino.runtime import Core
ie = Core()
model = ie.read_model(model="best_openvino_model/best.xml")
compiled_model = ie.compile_model(model=model, device_name="CPU")
input_layer_ir = compiled_model.input(0)
output_layer_ir = compiled_model.output()
image = cv2.imread("../test.jpg")
# N, C, H, W = input_layer_ir.shape
N, C, H, W = 1, 1, 480, 480
# resized_image = cv2.resize(image, (W, H))
# input_image = np.expand_dims(resized_image.transpose(2, 0, 1), 0)
input_image = cv2.dnn.blobFromImage(resized_image, 1/255, (W, H), [0,0,0], 1, crop=False)
output = compiled_model([input_image])[output_layer_ir]
output = cv2.transpose(output[0])
boxes = []
scores = []
class_ids = []
for row in output:
# Each row is [x, y, width, height, probability class 0, probability class 1, ...]
classes_scores = row[4:]
(_min_score, max_score, _min_class_loc, (_x, max_class_index)) = cv2.minMaxLoc(classes_scores)
if max_score >= 0.25:
box = [row[0] - (0.5 * row[2]), row[1] - (0.5 * row[3]), row[2], row[3]]
boxes.append(box)
scores.append(max_score)
class_ids.append(max_class_index)
[height, width, _] = image.shape
length = max((height, width))
scale = length/480
# Apply NMS (Non-maximum suppression)
RESULT_BOXES = cv2.dnn.NMSBoxes(boxes, scores, 0.5, 0.6, 0.5)
for index in RESULT_BOXES:
box = boxes[index]
x, y = round(box[0] * scale), round(box[1] * scale)
x_plus_w, y_plus_h = round((box[0] + box[2]) * scale), round((box[1] + box[3]) * scale)
image = cv2.rectangle(cv2.UMat(image), (x, y), (x_plus_w, y_plus_h), (0, 0, 255), 8)
image = cv2.putText(cv2.UMat(image), f"handgun {round(scores[index], 2)}", (x - 10, y - 12),
cv2.FONT_HERSHEY_SIMPLEX, 3, (0,0,255), 8)
makedirs("runs/openvino", exist_ok=True)
cv2.imwrite("test_openvino_runtime.jpg", image)
```
A bit late to the party, but I have stumbled upon this problem too, I believe from the output shape [1,6,8400] you are seeing you can take the first value and transpose to [8400, 6] giving you 8400 rows of the format [x, y, width, height, probability class 0, probability class 1], assuming you have 2 classes. So to process the code would look like:
```
import cv2
import matplotlib.pyplot as plt
import numpy as np
from openvino.runtime import Core
ie = Core()
model = ie.read_model(model=r"\models\IR\model_fit.xml")
compiled_model = ie.compile_model(model=model, device_name="CPU")
input_layer_ir = compiled_model.input(0)
output_layer_ir = compiled_model.output()
image = cv2.imread(r'\Resources\test_image.jpg')
N, C, H, W = input_layer_ir.shape
resized_image = cv2.resize(image, (W, H))
input_image = np.expand_dims(resized_image.transpose(2, 0, 1), 0)
plt.imshow(cv2.cvtColor(image, cv2.COLOR_BGR2RGB));
output = compiled_model([input_image])[output_layer_ir]
output = cv2.transpose(output[0])
print("Response shape", output.shape)
boxes = []
scores = []
class_ids = []
for row in output:
# Each row is [x, y, width, height, probability class 0, probability class 1, ...]
classes_scores = row[4:]
(_min_score, max_score, _min_class_loc, (_x, max_class_index)) = cv2.minMaxLoc(classes_scores)
if max_score >= 0.25:
box = [row[0] - (0.5 * row[2]), row[1] - (0.5 * row[3]), row[2], row[3]]
boxes.append(box)
scores.append(max_score)
class_ids.append(max_class_index)
[height, width, _] = image.shape
length = max((height, width))
scale = length/480
# Apply NMS (Non-maximum suppression)
RESULT_BOXES = cv2.dnn.NMSBoxes(boxes, scores, 0.5, 0.6, 0.5)
for index in RESULT_BOXES:
box = boxes[index]
x, y = round(box[0] * scale), round(box[1] * scale)
x_plus_w, y_plus_h = round((box[0] + box[2]) * scale), round((box[1] + box[3]) * scale)
draw_image = cv2.rectangle(cv2.UMat(image), (x, y), (x_plus_w, y_plus_h), (0, 0, 255), 4)
cv2.imwrite(r'\Resources\test_image_result.jpg', image)
```
There is more detail here: https://github.com/ultralytics/ultralytics/blob/main/examples/YOLOv8-OpenCV-ONNX-Python/main.py
Hope that helps, and if so could you let me know how you exported your model from YoloV8? I have done it but it gives me weird results using the IE Core instead of loading and running with YOLO
