Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -3,10 +3,10 @@ import cv2
|
|
3 |
import numpy as np
|
4 |
from ultralytics import YOLO
|
5 |
|
6 |
-
#
|
7 |
-
model = YOLO('yolo11s-earth.pt') #
|
8 |
|
9 |
-
#
|
10 |
default_classes = [
|
11 |
'airplane', 'airport', 'baseballfield', 'basketballcourt', 'bridge',
|
12 |
'chimney', 'dam', 'Expressway-Service-area', 'Expressway-toll-station',
|
@@ -16,41 +16,62 @@ default_classes = [
|
|
16 |
]
|
17 |
|
18 |
def process_frame(frame, classes_input):
|
19 |
-
#
|
20 |
-
if classes_input:
|
21 |
classes_list = [cls.strip() for cls in classes_input.split(',')]
|
22 |
-
|
|
|
|
|
|
|
|
|
|
|
23 |
else:
|
24 |
-
#
|
25 |
model.set_classes(default_classes)
|
26 |
|
27 |
-
#
|
28 |
frame = frame.copy()
|
29 |
|
30 |
-
#
|
31 |
-
|
|
|
|
|
32 |
|
33 |
-
#
|
34 |
-
|
35 |
|
36 |
-
#
|
|
|
|
|
|
|
37 |
for result in results:
|
38 |
boxes = result.boxes
|
39 |
for box in boxes:
|
40 |
x1, y1, x2, y2 = box.xyxy[0]
|
41 |
conf = box.conf[0]
|
42 |
cls = box.cls[0]
|
43 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
44 |
|
45 |
-
#
|
46 |
-
cv2.rectangle(frame, (
|
47 |
-
cv2.putText(frame, f'{class_name}:{conf:.2f}', (
|
48 |
|
49 |
return frame
|
50 |
|
51 |
def main():
|
52 |
-
#
|
53 |
with gr.Blocks() as demo:
|
|
|
54 |
with gr.Row():
|
55 |
cam_input = gr.Image(type="numpy", sources=["webcam"], streaming=True, label="Webcam")
|
56 |
classes_input = gr.Textbox(label="New classes (comma-separated)", placeholder="e.g.: airplane, airport, tennis court")
|
@@ -62,7 +83,7 @@ def main():
|
|
62 |
outputs=output
|
63 |
)
|
64 |
|
65 |
-
#
|
66 |
demo.launch()
|
67 |
|
68 |
if __name__ == "__main__":
|
|
|
3 |
import numpy as np
|
4 |
from ultralytics import YOLO
|
5 |
|
6 |
+
# Load YOLO model
|
7 |
+
model = YOLO('yolo11s-earth.pt') # Load your model
|
8 |
|
9 |
+
# Default classes
|
10 |
default_classes = [
|
11 |
'airplane', 'airport', 'baseballfield', 'basketballcourt', 'bridge',
|
12 |
'chimney', 'dam', 'Expressway-Service-area', 'Expressway-toll-station',
|
|
|
16 |
]
|
17 |
|
18 |
def process_frame(frame, classes_input):
|
19 |
+
# Process user input classes
|
20 |
+
if classes_input and classes_input.strip():
|
21 |
classes_list = [cls.strip() for cls in classes_input.split(',')]
|
22 |
+
# Validate classes_list
|
23 |
+
for cls in classes_list:
|
24 |
+
if not isinstance(cls, str):
|
25 |
+
print("Invalid class name:", cls)
|
26 |
+
continue
|
27 |
+
model.set_classes(classes_list) # Set model classes
|
28 |
else:
|
29 |
+
# Use default classes if no input or input is empty
|
30 |
model.set_classes(default_classes)
|
31 |
|
32 |
+
# Copy frame to a writable array
|
33 |
frame = frame.copy()
|
34 |
|
35 |
+
# Resize image to speed up processing (optional)
|
36 |
+
h, w = frame.shape[:2]
|
37 |
+
new_size = (640, int(h * (640 / w))) if w > h else (int(w * (640 / h)), 640)
|
38 |
+
resized_frame = cv2.resize(frame, new_size)
|
39 |
|
40 |
+
# Convert image format
|
41 |
+
rgb_frame = cv2.cvtColor(resized_frame, cv2.COLOR_BGR2RGB)
|
42 |
|
43 |
+
# Use model for detection
|
44 |
+
results = model.predict(rgb_frame)
|
45 |
+
|
46 |
+
# Draw detection results
|
47 |
for result in results:
|
48 |
boxes = result.boxes
|
49 |
for box in boxes:
|
50 |
x1, y1, x2, y2 = box.xyxy[0]
|
51 |
conf = box.conf[0]
|
52 |
cls = box.cls[0]
|
53 |
+
try:
|
54 |
+
class_name = model.names[int(cls)]
|
55 |
+
except (IndexError, TypeError) as e:
|
56 |
+
print(f"Error accessing model.names: {e}")
|
57 |
+
class_name = "Unknown" # Provide a default value
|
58 |
+
|
59 |
+
# Adjust coordinates to original image size
|
60 |
+
x1 = int(x1 * w / new_size[0])
|
61 |
+
y1 = int(y1 * h / new_size[1])
|
62 |
+
x2 = int(x2 * w / new_size[0])
|
63 |
+
y2 = int(y2 * h / new_size[1])
|
64 |
|
65 |
+
# Draw bounding box and label
|
66 |
+
cv2.rectangle(frame, (x1, y1), (x2, y2), (0, 255, 0), 2)
|
67 |
+
cv2.putText(frame, f'{class_name}:{conf:.2f}', (x1, y1 - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.9, (36, 255, 12), 2)
|
68 |
|
69 |
return frame
|
70 |
|
71 |
def main():
|
72 |
+
# Create Gradio interface
|
73 |
with gr.Blocks() as demo:
|
74 |
+
gr.Markdown("# YOLO11s-Earth open vocabulary detection (DIOR finetuning)")
|
75 |
with gr.Row():
|
76 |
cam_input = gr.Image(type="numpy", sources=["webcam"], streaming=True, label="Webcam")
|
77 |
classes_input = gr.Textbox(label="New classes (comma-separated)", placeholder="e.g.: airplane, airport, tennis court")
|
|
|
83 |
outputs=output
|
84 |
)
|
85 |
|
86 |
+
# Launch Gradio app
|
87 |
demo.launch()
|
88 |
|
89 |
if __name__ == "__main__":
|