Update app.py
Browse files
app.py
CHANGED
|
@@ -6,7 +6,7 @@ import numpy as np
|
|
| 6 |
from transformers import SamModel, SamProcessor
|
| 7 |
|
| 8 |
|
| 9 |
-
model = SamModel.from_pretrained('facebook/sam-vit-huge')
|
| 10 |
processor = SamProcessor.from_pretrained('facebook/sam-vit-huge')
|
| 11 |
|
| 12 |
|
|
@@ -14,8 +14,7 @@ def set_predictor(image):
|
|
| 14 |
"""
|
| 15 |
Creates a Sam predictor object based on a given image and model.
|
| 16 |
"""
|
| 17 |
-
|
| 18 |
-
inputs = processor(image, return_tensors='pt').to(device)
|
| 19 |
image_embedding = model.get_image_embeddings(inputs['pixel_values'])
|
| 20 |
|
| 21 |
return [image, image_embedding, 'Done']
|
|
@@ -28,8 +27,7 @@ def get_polygon(points, image, image_embedding):
|
|
| 28 |
"""
|
| 29 |
points = [int(w) for w in points.split(',')]
|
| 30 |
|
| 31 |
-
|
| 32 |
-
inputs = processor(image, input_boxes=[points], return_tensors="pt").to(device)
|
| 33 |
|
| 34 |
# pop the pixel_values as they are not neded
|
| 35 |
inputs.pop("pixel_values", None)
|
|
@@ -49,8 +47,11 @@ def get_polygon(points, image, image_embedding):
|
|
| 49 |
img = mask.astype(np.uint8)[0]
|
| 50 |
|
| 51 |
contours, hierarchy = cv2.findContours(img, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
|
| 52 |
-
points = contours[0]
|
| 53 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 54 |
polygon = []
|
| 55 |
for point in points:
|
| 56 |
for x, y in point:
|
|
@@ -73,6 +74,7 @@ with gr.Blocks() as app:
|
|
| 73 |
with gr.Tab('Get points'):
|
| 74 |
bbox = gr.Textbox(label="bbox")
|
| 75 |
polygon = [gr.Textbox(label='Polygon')]
|
|
|
|
| 76 |
points_button = gr.Button('Send bounding box')
|
| 77 |
|
| 78 |
|
|
@@ -85,7 +87,7 @@ with gr.Blocks() as app:
|
|
| 85 |
points_button.click(
|
| 86 |
get_polygon,
|
| 87 |
[bbox, image, embedding],
|
| 88 |
-
polygon,
|
| 89 |
)
|
| 90 |
|
| 91 |
app.launch(debug=True)
|
|
|
|
| 6 |
from transformers import SamModel, SamProcessor
|
| 7 |
|
| 8 |
|
| 9 |
+
model = SamModel.from_pretrained('facebook/sam-vit-huge').to('cuda')
|
| 10 |
processor = SamProcessor.from_pretrained('facebook/sam-vit-huge')
|
| 11 |
|
| 12 |
|
|
|
|
| 14 |
"""
|
| 15 |
Creates a Sam predictor object based on a given image and model.
|
| 16 |
"""
|
| 17 |
+
inputs = processor(image, return_tensors='pt').to('cuda')
|
|
|
|
| 18 |
image_embedding = model.get_image_embeddings(inputs['pixel_values'])
|
| 19 |
|
| 20 |
return [image, image_embedding, 'Done']
|
|
|
|
| 27 |
"""
|
| 28 |
points = [int(w) for w in points.split(',')]
|
| 29 |
|
| 30 |
+
inputs = processor(image, input_boxes=[points], return_tensors="pt").to('cuda')
|
|
|
|
| 31 |
|
| 32 |
# pop the pixel_values as they are not neded
|
| 33 |
inputs.pop("pixel_values", None)
|
|
|
|
| 47 |
img = mask.astype(np.uint8)[0]
|
| 48 |
|
| 49 |
contours, hierarchy = cv2.findContours(img, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
|
|
|
|
| 50 |
|
| 51 |
+
if len(contours) == 0:
|
| 52 |
+
return [0], img
|
| 53 |
+
|
| 54 |
+
points = contours[0]
|
| 55 |
polygon = []
|
| 56 |
for point in points:
|
| 57 |
for x, y in point:
|
|
|
|
| 74 |
with gr.Tab('Get points'):
|
| 75 |
bbox = gr.Textbox(label="bbox")
|
| 76 |
polygon = [gr.Textbox(label='Polygon')]
|
| 77 |
+
mask = gr.Image(label='Mask')
|
| 78 |
points_button = gr.Button('Send bounding box')
|
| 79 |
|
| 80 |
|
|
|
|
| 87 |
points_button.click(
|
| 88 |
get_polygon,
|
| 89 |
[bbox, image, embedding],
|
| 90 |
+
[polygon, mask],
|
| 91 |
)
|
| 92 |
|
| 93 |
app.launch(debug=True)
|