swenkel
commited on
Commit
·
4fcc8b2
1
Parent(s):
195fa4b
text_recognition example argparse bug fix (#89)
Browse files
demo.py
CHANGED
|
@@ -44,6 +44,10 @@ parser.add_argument('--target', '-t', type=int, default=targets[0], help=help_ms
|
|
| 44 |
parser.add_argument('--charset', '-c', type=str, default='charset_36_EN.txt', help='Path to the charset file corresponding to the selected model.')
|
| 45 |
parser.add_argument('--save', '-s', type=str, default=False, help='Set true to save results. This flag is invalid when using camera.')
|
| 46 |
parser.add_argument('--vis', '-v', type=str2bool, default=True, help='Set true to open a window for result visualization. This flag is invalid when using camera.')
|
|
|
|
|
|
|
|
|
|
|
|
|
| 47 |
args = parser.parse_args()
|
| 48 |
|
| 49 |
def visualize(image, boxes, texts, color=(0, 255, 0), isClosed=True, thickness=2):
|
|
@@ -60,7 +64,7 @@ if __name__ == '__main__':
|
|
| 60 |
recognizer = CRNN(modelPath=args.model, charsetPath=args.charset)
|
| 61 |
# Instantiate DB for text detection
|
| 62 |
detector = DB(modelPath='../text_detection_db/text_detection_DB_IC15_resnet18_2021sep.onnx',
|
| 63 |
-
inputSize=[
|
| 64 |
binaryThreshold=0.3,
|
| 65 |
polygonThreshold=0.5,
|
| 66 |
maxCandidates=200,
|
|
@@ -106,7 +110,7 @@ if __name__ == '__main__':
|
|
| 106 |
print('No frames grabbed!')
|
| 107 |
break
|
| 108 |
|
| 109 |
-
frame = cv.resize(frame, [
|
| 110 |
# Inference of text detector
|
| 111 |
tm.start()
|
| 112 |
results = detector.infer(frame)
|
|
|
|
| 44 |
parser.add_argument('--charset', '-c', type=str, default='charset_36_EN.txt', help='Path to the charset file corresponding to the selected model.')
|
| 45 |
parser.add_argument('--save', '-s', type=str, default=False, help='Set true to save results. This flag is invalid when using camera.')
|
| 46 |
parser.add_argument('--vis', '-v', type=str2bool, default=True, help='Set true to open a window for result visualization. This flag is invalid when using camera.')
|
| 47 |
+
parser.add_argument('--width', type=int, default=736,
|
| 48 |
+
help='Preprocess input image by resizing to a specific width. It should be multiple by 32.')
|
| 49 |
+
parser.add_argument('--height', type=int, default=736,
|
| 50 |
+
help='Preprocess input image by resizing to a specific height. It should be multiple by 32.')
|
| 51 |
args = parser.parse_args()
|
| 52 |
|
| 53 |
def visualize(image, boxes, texts, color=(0, 255, 0), isClosed=True, thickness=2):
|
|
|
|
| 64 |
recognizer = CRNN(modelPath=args.model, charsetPath=args.charset)
|
| 65 |
# Instantiate DB for text detection
|
| 66 |
detector = DB(modelPath='../text_detection_db/text_detection_DB_IC15_resnet18_2021sep.onnx',
|
| 67 |
+
inputSize=[args.width, args.height],
|
| 68 |
binaryThreshold=0.3,
|
| 69 |
polygonThreshold=0.5,
|
| 70 |
maxCandidates=200,
|
|
|
|
| 110 |
print('No frames grabbed!')
|
| 111 |
break
|
| 112 |
|
| 113 |
+
frame = cv.resize(frame, [args.width, args.height])
|
| 114 |
# Inference of text detector
|
| 115 |
tm.start()
|
| 116 |
results = detector.infer(frame)
|