yolo2voc
yolo2voc copied to clipboard
🔄 | YOLO2VOC - VOC2YOLO
Convert YOLO to Pascal VOC format or Pascal to YOLO format
Usage:
git clone https://github.com/yakhyo/yolo2voc.git
cd yolo2voc
pip install -r requirements.txt
- Modify variables in
config.py
based according to your dataset - Run python
main.py --yolo2voc
to convert YOLO to VOC - Run python
main.py --voc2yolo
to convert VOC to YOLO - Run python
main.py --voc2yolo_a
to convert VOC to YOLO (absolute)
Pascal VOC To YOLO
main.py --voc2yolo
-
<object-class>
- integer number of object from0
to(classes-1)
-
<x> <y> <width> <height>
-float
values relative to width and height of an image, it can be within range of(0.0 to 1.0]
- For example:
<x> = <absolute_x> / <image_width> or <height> = <absolute_height> / <image_height>
- Attention:
<x> <y>
- are center of rectangle (are not top-left corner)
zidane.jpg
The label file corresponding to the above image contains 2 persons (class 0) and a tie (class 27):
<object-class> <x> <y> <width> <height>
Pascal VOC To YOLO Absolute
main.py --voc2yolo_a
-
<object-class>
- integer number of object from0
to(classes-1)
-
<x_min> <y_min> <x_max> <y_max>
-int
absolute values object coordinates - For example:
<object-class> <x_min> <y_min> <x_max> <y_max>
:
1 255 247 425 468
0 470 105 680 468
1 152 356 658 754
Reference
- https://github.com/AlexeyAB/Yolo_mark/issues/60
- https://github.com/jahongir7174/YOLO2VOC