BalloonInc / VisDrone_YOLOv5

Building Custom Object Detection model on VisDrone2021 dataset

VisDrone_YOLOv5

Building Custom Object Detection model on VisDrone2021 dataset using state of the art YoloV5

Files:

1.convert.py: normalising cooridnates for bounding boxes
2.best.pt: best weight for our trained model
2.VisDrone_YOLOv5.ipynb: Training custom object detector for 11 classes that include:

{ pedestrian(1),people(2), bicycle(3), car(4), van(5), truck(6), tricycle(7), awning-tricycle(8), bus(9), motor(10),others(11) }

Results

Graphs

Train_batch images

Test Images