Real-Time-Object-Detection-using-yolo

Performance on the COCO Dataset

We Need only three files

• yolov3.weights
• colo-name
• yolov3.cfg

Download the coco-name and colo-name file here Or Just run this

$ git clone https://github.com/pjreddie/darknet

• coco-name location:- darknet/data/coco.names
• yolov3.cfg location:- darknet/cfg/yolov3.cfg

For Download the pre-trained weight

$ wget https://pjreddie.com/media/files/yolov3.weights

For Webcam Just use 0

If you want to Change confidence and threshold value

$ python3 yolo_video.py -i o -c 0.6 -t 0.4

• -c for confidence
• -t for threshold

For other Media file and live stream Just run this

$ python3 yolo_video.py -i rtsp://192.168.43.102:7878/h264_ulaw.sdp (use your own ip)

For live stream by url Just run this

$ python3 yolo_video.py -i http://192.168.43.102:7878/video?x.mjpg #(use your own ip)

output

For Video file

$ python3 yolo_video.py -i video_file_name