Alessio Grancini commited on
Commit
7021f6a
·
verified ·
1 Parent(s): f407854

Update image_segmenter.py

Browse files
Files changed (1) hide show
  1. image_segmenter.py +81 -43
image_segmenter.py CHANGED
@@ -3,10 +3,12 @@ import numpy as np
3
  from ultralytics import YOLO
4
  import random
5
  import spaces
 
 
6
 
7
  class ImageSegmenter:
8
  def __init__(self, model_type="yolov8s-seg") -> None:
9
- # Don't initialize any CUDA/GPU stuff here
10
  self.model_type = model_type
11
  self.is_show_bounding_boxes = True
12
  self.is_show_segmentation_boundary = False
@@ -17,6 +19,17 @@ class ImageSegmenter:
17
  self.bb_clr = (255, 0, 0)
18
  self.masks = {}
19
  self.model = None
 
 
 
 
 
 
 
 
 
 
 
20
 
21
  def get_cls_clr(self, cls_id):
22
  if cls_id in self.cls_clr:
@@ -29,59 +42,84 @@ class ImageSegmenter:
29
 
30
  @spaces.GPU
31
  def predict(self, image):
32
- # Initialize model if needed
33
- if self.model is None:
34
- print("Loading YOLO model...")
35
- self.model = YOLO('models/' + self.model_type + '.pt')
36
- print("Model loaded successfully")
 
 
 
 
37
 
38
- # params
39
- objects_data = []
40
- image = image.copy()
41
- predictions = self.model.predict(image)
 
42
 
43
- cls_ids = predictions[0].boxes.cls.cpu().numpy()
44
- bounding_boxes = predictions[0].boxes.xyxy.int().cpu().numpy()
45
- cls_conf = predictions[0].boxes.conf.cpu().numpy()
46
-
47
- if predictions[0].masks:
48
- seg_mask_boundary = predictions[0].masks.xy
49
- seg_mask = predictions[0].masks.data.cpu().numpy()
50
- else:
51
- seg_mask_boundary, seg_mask = [], np.array([])
52
-
53
- for id, cls in enumerate(cls_ids):
54
- cls_clr = self.get_cls_clr(cls)
55
 
56
- if seg_mask.any() and cls_conf[id] > self.confidence_threshold:
57
- self.masks[id] = seg_mask[id]
58
-
59
- if self.is_show_segmentation:
60
- alpha = 0.8
61
- colored_mask = np.expand_dims(seg_mask[id], 0).repeat(3, axis=0)
62
- colored_mask = np.moveaxis(colored_mask, 0, -1)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
63
 
64
- if image.shape[:2] != seg_mask[id].shape[:2]:
65
- colored_mask = cv2.resize(colored_mask, (image.shape[1], image.shape[0]))
66
 
67
- masked = np.ma.MaskedArray(image, mask=colored_mask, fill_value=cls_clr)
68
- image_overlay = masked.filled()
69
- image = cv2.addWeighted(image, 1 - alpha, image_overlay, alpha, 0)
70
 
71
- if self.is_show_bounding_boxes and cls_conf[id] > self.confidence_threshold:
72
  (x1, y1, x2, y2) = bounding_boxes[id]
73
- cls_name = self.model.names[cls]
74
  cls_confidence = cls_conf[id]
75
- disp_str = cls_name +' '+ str(round(cls_confidence, 2))
76
  cv2.rectangle(image, (x1, y1), (x2, y2), cls_clr, self.bb_thickness)
77
- cv2.rectangle(image, (x1, y1), (x1+(len(disp_str)*9), y1+15), cls_clr, -1)
78
  cv2.putText(image, disp_str, (x1+5, y1+10), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (255, 255, 255), 1)
79
 
80
- if len(seg_mask_boundary) and self.is_show_segmentation_boundary and cls_conf[id] > self.confidence_threshold:
81
- cv2.polylines(image, [np.array(seg_mask_boundary[id], dtype=np.int32)], isClosed=True, color=cls_clr, thickness=2)
 
82
 
83
  (x1, y1, x2, y2) = bounding_boxes[id]
84
- center = x1+(x2-x1)//2, y1+(y2-y1)//2
85
- objects_data.append([cls, self.model.names[cls], center, self.masks[id], cls_clr])
 
86
 
87
- return image, objects_data
 
 
 
 
 
 
 
3
  from ultralytics import YOLO
4
  import random
5
  import spaces
6
+ import os
7
+ import torch
8
 
9
  class ImageSegmenter:
10
  def __init__(self, model_type="yolov8s-seg") -> None:
11
+ # Initialize parameters
12
  self.model_type = model_type
13
  self.is_show_bounding_boxes = True
14
  self.is_show_segmentation_boundary = False
 
19
  self.bb_clr = (255, 0, 0)
20
  self.masks = {}
21
  self.model = None
22
+
23
+ # Ensure model directory exists
24
+ os.makedirs('models', exist_ok=True)
25
+
26
+ # Check if model file exists, if not download it
27
+ model_path = os.path.join('models', f'{model_type}.pt')
28
+ if not os.path.exists(model_path):
29
+ print(f"Downloading {model_type} model...")
30
+ self.model = YOLO(model_type)
31
+ self.model.export()
32
+ print("Model downloaded successfully")
33
 
34
  def get_cls_clr(self, cls_id):
35
  if cls_id in self.cls_clr:
 
42
 
43
  @spaces.GPU
44
  def predict(self, image):
45
+ try:
46
+ # Initialize model if needed
47
+ if self.model is None:
48
+ print("Loading YOLO model...")
49
+ model_path = os.path.join('models', f'{self.model_type}.pt')
50
+ # Force CPU mode for YOLO initialization
51
+ self.model = YOLO(model_path)
52
+ self.model.to('cpu') # Explicitly move to CPU
53
+ print("Model loaded successfully")
54
 
55
+ # Ensure image is in correct format
56
+ if isinstance(image, np.ndarray):
57
+ image = image.copy()
58
+ else:
59
+ raise ValueError("Input image must be a numpy array")
60
 
61
+ # Make prediction using CPU
62
+ predictions = self.model.predict(image, device='cpu')
 
 
 
 
 
 
 
 
 
 
63
 
64
+ # Process results
65
+ objects_data = []
66
+
67
+ if len(predictions) == 0 or not predictions[0].boxes:
68
+ return image, objects_data
69
+
70
+ cls_ids = predictions[0].boxes.cls.numpy() # Changed from cpu().numpy()
71
+ bounding_boxes = predictions[0].boxes.xyxy.int().numpy()
72
+ cls_conf = predictions[0].boxes.conf.numpy()
73
+
74
+ if predictions[0].masks is not None:
75
+ seg_mask_boundary = predictions[0].masks.xy
76
+ seg_mask = predictions[0].masks.data.numpy() # Changed from cpu().numpy()
77
+ else:
78
+ seg_mask_boundary, seg_mask = [], np.array([])
79
+
80
+ for id, cls in enumerate(cls_ids):
81
+ if cls_conf[id] <= self.confidence_threshold:
82
+ continue
83
+
84
+ cls_clr = self.get_cls_clr(int(cls))
85
+
86
+ if seg_mask.size > 0:
87
+ self.masks[id] = seg_mask[id]
88
+
89
+ if self.is_show_segmentation:
90
+ alpha = 0.8
91
+ colored_mask = np.expand_dims(seg_mask[id], 0).repeat(3, axis=0)
92
+ colored_mask = np.moveaxis(colored_mask, 0, -1)
93
 
94
+ if image.shape[:2] != seg_mask[id].shape[:2]:
95
+ colored_mask = cv2.resize(colored_mask, (image.shape[1], image.shape[0]))
96
 
97
+ masked = np.ma.MaskedArray(image, mask=colored_mask, fill_value=cls_clr)
98
+ image_overlay = masked.filled()
99
+ image = cv2.addWeighted(image, 1 - alpha, image_overlay, alpha, 0)
100
 
101
+ if self.is_show_bounding_boxes:
102
  (x1, y1, x2, y2) = bounding_boxes[id]
103
+ cls_name = self.model.names[int(cls)]
104
  cls_confidence = cls_conf[id]
105
+ disp_str = f"{cls_name} {cls_confidence:.2f}"
106
  cv2.rectangle(image, (x1, y1), (x2, y2), cls_clr, self.bb_thickness)
107
+ cv2.rectangle(image, (x1, y1), (x1+len(disp_str)*9, y1+15), cls_clr, -1)
108
  cv2.putText(image, disp_str, (x1+5, y1+10), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (255, 255, 255), 1)
109
 
110
+ if len(seg_mask_boundary) > 0 and self.is_show_segmentation_boundary:
111
+ cv2.polylines(image, [np.array(seg_mask_boundary[id], dtype=np.int32)],
112
+ isClosed=True, color=cls_clr, thickness=2)
113
 
114
  (x1, y1, x2, y2) = bounding_boxes[id]
115
+ center = (x1+(x2-x1)//2, y1+(y2-y1)//2)
116
+ objects_data.append([int(cls), self.model.names[int(cls)], center,
117
+ self.masks.get(id, None), cls_clr])
118
 
119
+ return image, objects_data
120
+
121
+ except Exception as e:
122
+ print(f"Error in predict: {str(e)}")
123
+ import traceback
124
+ print(traceback.format_exc())
125
+ raise