/box_yolov8

Yolo v8 model for box detection.

Primary LanguageJupyter NotebookMIT LicenseMIT

box_yolov8

Yolo v8s (Small) model for box detection.

Instructions for use

  • Clone the repository
git clone https://github.com/Vamsi-IITI/box_yolov8.git
  • Place test images in folder box_yolov8/test/images
  • Run
yolo task=detect mode=predict model=~/box_yolov8/best.pt conf=0.25 source=~/box_yolov8/test/images save=True

For detection in image

  • Use count.py . Run
python3 count.py

For object detection and tracking in video

  • Use track.py . Mention the video in the code. Run
python3 track.py
2023-07-13.16-10-51.1.mp4

Video Demo

Note :

Both track_and_video_save.py and box_detection_ros.py have some issues currently

Training Results :

image

Few Results :

two_boxes

cardboardbox

image