This is from scratch implementation of yolov2 model. I have taken help from fairyonices blog https://fairyonice.github.io/Part_1_Object_Detection_with_Yolo_for_VOC_2014_data_anchor_box_clustering.html
I have used tensorflow 2.2 and also some dependent python libraries like matplotlib and json. I have referred the yolov2 and yolov3 papers while creating this project. The model is overfit on 100 images from the MS COCO train 2014 dataset, I have done so to just check if the model works as intended. I also have taken help from aladinn perssons youtube channel as well as github repo.
Current output on the overfit data itself:
The papers:
https://arxiv.org/pdf/1804.02767 - YOLOv3
https://arxiv.org/pdf/1612.08242 - YOLO9000