Update app.py
Browse files
app.py
CHANGED
@@ -1,9 +1,9 @@
|
|
1 |
import gradio as gr
|
2 |
import sahi
|
3 |
import torch
|
4 |
-
from ultralyticsplus import YOLO
|
5 |
|
6 |
-
#
|
7 |
sahi.utils.file.download_from_url(
|
8 |
"https://raw.githubusercontent.com/kadirnar/dethub/main/data/images/highway.jpg",
|
9 |
"highway.jpg",
|
@@ -17,7 +17,7 @@ sahi.utils.file.download_from_url(
|
|
17 |
"zidane.jpg",
|
18 |
)
|
19 |
|
20 |
-
|
21 |
model_names = [
|
22 |
"yolov8n-seg.pt",
|
23 |
"yolov8s-seg.pt",
|
@@ -26,10 +26,10 @@ model_names = [
|
|
26 |
"yolov8x-seg.pt",
|
27 |
]
|
28 |
|
|
|
29 |
current_model_name = "yolov8m-seg.pt"
|
30 |
model = YOLO(current_model_name)
|
31 |
|
32 |
-
|
33 |
def yolov8_inference(
|
34 |
image: gr.Image = None,
|
35 |
model_name: gr.Dropdown = None,
|
@@ -38,7 +38,7 @@ def yolov8_inference(
|
|
38 |
iou_threshold: gr.Slider = 0.45,
|
39 |
):
|
40 |
"""
|
41 |
-
YOLOv8 inference function
|
42 |
Args:
|
43 |
image: Input image
|
44 |
model_name: Name of the model
|
@@ -46,29 +46,34 @@ def yolov8_inference(
|
|
46 |
conf_threshold: Confidence threshold
|
47 |
iou_threshold: IOU threshold
|
48 |
Returns:
|
49 |
-
|
50 |
"""
|
51 |
global model
|
52 |
global current_model_name
|
|
|
|
|
53 |
if model_name != current_model_name:
|
54 |
model = YOLO(model_name)
|
55 |
current_model_name = model_name
|
|
|
|
|
56 |
model.overrides["conf"] = conf_threshold
|
57 |
model.overrides["iou"] = iou_threshold
|
|
|
|
|
58 |
results = model.predict(image, imgsz=image_size, return_outputs=True)
|
59 |
-
|
60 |
-
|
61 |
-
|
62 |
-
|
63 |
-
|
64 |
-
|
65 |
-
|
66 |
-
|
67 |
-
|
68 |
-
|
69 |
-
return masks
|
70 |
-
|
71 |
|
|
|
72 |
inputs = [
|
73 |
gr.Image(type="filepath", label="Input Image"),
|
74 |
gr.Dropdown(
|
@@ -83,14 +88,18 @@ inputs = [
|
|
83 |
gr.Slider(minimum=0.0, maximum=1.0, value=0.45, step=0.05, label="IOU Threshold"),
|
84 |
]
|
85 |
|
86 |
-
|
87 |
-
|
88 |
|
|
|
|
|
89 |
examples = [
|
90 |
["zidane.jpg", "yolov8m-seg.pt", 640, 0.6, 0.45],
|
91 |
["highway.jpg", "yolov8m-seg.pt", 640, 0.25, 0.45],
|
92 |
["small-vehicles1.jpeg", "yolov8m-seg.pt", 640, 0.25, 0.45],
|
93 |
]
|
|
|
|
|
94 |
demo_app = gr.Interface(
|
95 |
fn=yolov8_inference,
|
96 |
inputs=inputs,
|
@@ -100,4 +109,6 @@ demo_app = gr.Interface(
|
|
100 |
cache_examples=True,
|
101 |
theme="default",
|
102 |
)
|
103 |
-
|
|
|
|
|
|
1 |
import gradio as gr
|
2 |
import sahi
|
3 |
import torch
|
4 |
+
from ultralyticsplus import YOLO
|
5 |
|
6 |
+
# Download example images
|
7 |
sahi.utils.file.download_from_url(
|
8 |
"https://raw.githubusercontent.com/kadirnar/dethub/main/data/images/highway.jpg",
|
9 |
"highway.jpg",
|
|
|
17 |
"zidane.jpg",
|
18 |
)
|
19 |
|
20 |
+
# Define the model names
|
21 |
model_names = [
|
22 |
"yolov8n-seg.pt",
|
23 |
"yolov8s-seg.pt",
|
|
|
26 |
"yolov8x-seg.pt",
|
27 |
]
|
28 |
|
29 |
+
# Set the initial model
|
30 |
current_model_name = "yolov8m-seg.pt"
|
31 |
model = YOLO(current_model_name)
|
32 |
|
|
|
33 |
def yolov8_inference(
|
34 |
image: gr.Image = None,
|
35 |
model_name: gr.Dropdown = None,
|
|
|
38 |
iou_threshold: gr.Slider = 0.45,
|
39 |
):
|
40 |
"""
|
41 |
+
YOLOv8 inference function to return masks and label names for each detected object
|
42 |
Args:
|
43 |
image: Input image
|
44 |
model_name: Name of the model
|
|
|
46 |
conf_threshold: Confidence threshold
|
47 |
iou_threshold: IOU threshold
|
48 |
Returns:
|
49 |
+
Object masks, coordinates, and label names
|
50 |
"""
|
51 |
global model
|
52 |
global current_model_name
|
53 |
+
|
54 |
+
# Check if a new model is selected
|
55 |
if model_name != current_model_name:
|
56 |
model = YOLO(model_name)
|
57 |
current_model_name = model_name
|
58 |
+
|
59 |
+
# Set the confidence and IOU thresholds
|
60 |
model.overrides["conf"] = conf_threshold
|
61 |
model.overrides["iou"] = iou_threshold
|
62 |
+
|
63 |
+
# Perform model prediction
|
64 |
results = model.predict(image, imgsz=image_size, return_outputs=True)
|
65 |
+
|
66 |
+
# Extract masks, coordinates, and label names
|
67 |
+
output = []
|
68 |
+
for result in results:
|
69 |
+
for mask, box in zip(result.masks.xy, result.boxes):
|
70 |
+
label = model.names[int(box.cls[0])]
|
71 |
+
mask_coords = mask.tolist() # Convert mask coordinates to list format
|
72 |
+
output.append({"label": label, "mask_coords": mask_coords})
|
73 |
+
|
74 |
+
return output
|
|
|
|
|
75 |
|
76 |
+
# Define Gradio input and output components
|
77 |
inputs = [
|
78 |
gr.Image(type="filepath", label="Input Image"),
|
79 |
gr.Dropdown(
|
|
|
88 |
gr.Slider(minimum=0.0, maximum=1.0, value=0.45, step=0.05, label="IOU Threshold"),
|
89 |
]
|
90 |
|
91 |
+
# Output the object masks and label names as a JSON-like format
|
92 |
+
outputs = gr.JSON(label="Detected Objects with Masks and Labels")
|
93 |
|
94 |
+
# Title and example inputs
|
95 |
+
title = "Ultralytics YOLOv8 Segmentation Demo"
|
96 |
examples = [
|
97 |
["zidane.jpg", "yolov8m-seg.pt", 640, 0.6, 0.45],
|
98 |
["highway.jpg", "yolov8m-seg.pt", 640, 0.25, 0.45],
|
99 |
["small-vehicles1.jpeg", "yolov8m-seg.pt", 640, 0.25, 0.45],
|
100 |
]
|
101 |
+
|
102 |
+
# Create the Gradio interface
|
103 |
demo_app = gr.Interface(
|
104 |
fn=yolov8_inference,
|
105 |
inputs=inputs,
|
|
|
109 |
cache_examples=True,
|
110 |
theme="default",
|
111 |
)
|
112 |
+
|
113 |
+
# Launch the app
|
114 |
+
demo_app.queue().launch(debug=True)
|