search

hukaixuan19970627 / yolov5_obb

yolov5 + csl_label.(Oriented Object Detection)（Rotation Detection）（Rotated BBox）基于yolov5的旋转目标检测
GNU General Public License v3.0
1.79k stars 421 forks source link

Annotating images #552

Open GattupalliMonica opened 1 year ago

GattupalliMonica commented 1 year ago

How can we annotate the images to train in YOLOv5_obb. I means should we need to create rectangular bounding box and then convert them to yolov5_obb format?

ibtehajali67 commented 1 year ago

you have to find the four corner point with x,y and label_name and class no. As you can see in sample image and txt file

you can also check that issue https://github.com/hukaixuan19970627/yolov5_obb/issues/320

1.txt 1

RayOct18 commented 1 year ago

You can use this tool https://github.com/RayOct18/labelImg.

I have integrated the code from cgvict into the current version of LabelImg, which allows for directly saving txt in the YOLOv5 OBB format needed for training.

After following the installation instructions in the README file, open the LabelImg program and click the button below "Save" until "RotatedYOLO" appears, and then you can start annotating!