Spaces:
Running
on
T4
Running
on
T4
AAAAAAyq
commited on
Commit
·
87c6f54
1
Parent(s):
4d26566
Update application file
Browse files
app.py
CHANGED
@@ -4,11 +4,30 @@ import numpy as np
|
|
4 |
import matplotlib.pyplot as plt
|
5 |
import gradio as gr
|
6 |
import io
|
|
|
7 |
# import cv2
|
8 |
|
9 |
model = YOLO('checkpoints/FastSAM.pt') # load a custom model
|
10 |
|
11 |
-
def
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
12 |
if random_color : # random mask color
|
13 |
color = np.concatenate([np.random.random(3), np.array([0.6])], axis=0)
|
14 |
else:
|
@@ -28,28 +47,27 @@ def show_mask(annotation, ax, random_color=False, bbox=None, points=None):
|
|
28 |
ax.imshow(mask_image)
|
29 |
return mask_image
|
30 |
|
31 |
-
def post_process(annotations, image, mask_random_color=
|
32 |
-
# image = cv2.imread(image_path)
|
33 |
-
# image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
|
34 |
plt.figure(figsize=(10, 10))
|
35 |
plt.imshow(image)
|
36 |
for i, mask in enumerate(annotations):
|
37 |
show_mask(mask, plt.gca(),random_color=mask_random_color,bbox=bbox,points=points)
|
38 |
plt.axis('off')
|
39 |
-
|
40 |
# create a BytesIO object
|
41 |
buf = io.BytesIO()
|
42 |
|
43 |
# save plot to buf
|
44 |
plt.savefig(buf, format='png', bbox_inches='tight', pad_inches=0.0)
|
45 |
-
# plt.savefig('buffer/tmp.png', bbox_inches='tight', pad_inches=0.0)
|
46 |
|
47 |
# use PIL to open the image
|
48 |
img = Image.open(buf)
|
49 |
|
|
|
|
|
|
|
50 |
# don't forget to close the buffer
|
51 |
buf.close()
|
52 |
-
return
|
53 |
|
54 |
|
55 |
# def show_mask(annotation, ax, random_color=False):
|
@@ -77,10 +95,15 @@ def post_process(annotations, image, mask_random_color=False, bbox=None, points=
|
|
77 |
# post_process(results[0].masks, Image.open("../data/cake.png"))
|
78 |
|
79 |
def predict(inp):
|
80 |
-
results = model(inp, device='
|
81 |
-
|
|
|
82 |
return pil_image
|
83 |
|
|
|
|
|
|
|
|
|
84 |
|
85 |
demo = gr.Interface(fn=predict,
|
86 |
inputs=gr.inputs.Image(type='pil'),
|
|
|
4 |
import matplotlib.pyplot as plt
|
5 |
import gradio as gr
|
6 |
import io
|
7 |
+
import torch
|
8 |
# import cv2
|
9 |
|
10 |
model = YOLO('checkpoints/FastSAM.pt') # load a custom model
|
11 |
|
12 |
+
def format_results(result,filter = 0):
|
13 |
+
annotations = []
|
14 |
+
n = len(result.masks.data)
|
15 |
+
for i in range(n):
|
16 |
+
annotation = {}
|
17 |
+
mask = result.masks.data[i] == 1.0
|
18 |
+
|
19 |
+
|
20 |
+
if torch.sum(mask) < filter:
|
21 |
+
continue
|
22 |
+
annotation['id'] = i
|
23 |
+
annotation['segmentation'] = mask.cpu().numpy()
|
24 |
+
annotation['bbox'] = result.boxes.data[i]
|
25 |
+
annotation['score'] = result.boxes.conf[i]
|
26 |
+
annotation['area'] = annotation['segmentation'].sum()
|
27 |
+
annotations.append(annotation)
|
28 |
+
return annotations
|
29 |
+
|
30 |
+
def show_mask(annotation, ax, random_color=True, bbox=None, points=None):
|
31 |
if random_color : # random mask color
|
32 |
color = np.concatenate([np.random.random(3), np.array([0.6])], axis=0)
|
33 |
else:
|
|
|
47 |
ax.imshow(mask_image)
|
48 |
return mask_image
|
49 |
|
50 |
+
def post_process(annotations, image, mask_random_color=True, bbox=None, points=None):
|
|
|
|
|
51 |
plt.figure(figsize=(10, 10))
|
52 |
plt.imshow(image)
|
53 |
for i, mask in enumerate(annotations):
|
54 |
show_mask(mask, plt.gca(),random_color=mask_random_color,bbox=bbox,points=points)
|
55 |
plt.axis('off')
|
|
|
56 |
# create a BytesIO object
|
57 |
buf = io.BytesIO()
|
58 |
|
59 |
# save plot to buf
|
60 |
plt.savefig(buf, format='png', bbox_inches='tight', pad_inches=0.0)
|
|
|
61 |
|
62 |
# use PIL to open the image
|
63 |
img = Image.open(buf)
|
64 |
|
65 |
+
# copy the image data
|
66 |
+
img_copy = img.copy()
|
67 |
+
|
68 |
# don't forget to close the buffer
|
69 |
buf.close()
|
70 |
+
return img_copy
|
71 |
|
72 |
|
73 |
# def show_mask(annotation, ax, random_color=False):
|
|
|
95 |
# post_process(results[0].masks, Image.open("../data/cake.png"))
|
96 |
|
97 |
def predict(inp):
|
98 |
+
results = model(inp, device='cpu', retina_masks=True, iou=0.7, conf=0.25, imgsz=1024)
|
99 |
+
results = format_results(results[0], 100)
|
100 |
+
pil_image = post_process(annotations=results, image=inp)
|
101 |
return pil_image
|
102 |
|
103 |
+
# inp = 'assets/sa_192.jpg'
|
104 |
+
# results = model(inp, device='cpu', retina_masks=True, iou=0.7, conf=0.25, imgsz=1024)
|
105 |
+
# results = format_results(results[0], 100)
|
106 |
+
# post_process(annotations=results, image_path=inp)
|
107 |
|
108 |
demo = gr.Interface(fn=predict,
|
109 |
inputs=gr.inputs.Image(type='pil'),
|