Real-time Object Detection: YOLOv1 Re-Implementation in PyTorch

TL;DR

This paper presents a PyTorch implementation of YOLOv1 for real-time object detection, exploring modifications to improve performance and comparing results with the original architecture.

Contribution

It provides a re-implementation of YOLOv1 in PyTorch and investigates architectural modifications to enhance detection accuracy.

Findings

Modified architecture improved detection metrics

PyTorch implementation achieved real-time performance

Comparison shows competitive results with original YOLOv1

Abstract

Real-time object detection is a crucial problem to solve when in comes to computer vision systems that needs to make appropriate decision based on detection in a timely manner. I have chosen the YOLO v1 architecture to implement it using PyTorch framework, with goal to familiarize with entire object detection pipeline I attempted different techniques to modify the original architecture to improve the results. Finally, I compare the metrics of my implementation to the original.