YOLO Nano implementation

This is my implementation of YOLO Nano with Pytorch. If you like it, please star it. This repository could be freely used for any purpose. If you like it, please star it.

The code is not finished. Still ongoing

Introduction

YOLO Nano paper: YOLO Nano: a Highly Compact You Only Look Once Convolutional Neural Network for Object Detection

The network structure of YOLO Nano:

The performance of YOLO Nano:

How to run this code base

0. make sure you have powerfull GPU computer and you are good at Python coding.
1. git clone [email protected]:ardeal/yolo_nano.git
2. install Pytorch and necessary packages
3. prepare validating images ......

In addition

1. the model file

Time needed to run the code on CPU

On Windows 10, Intel(R) Xeon(R) W-2123 CPU @ 3.60GHz, 64G memory and SSD, the time needed to run:
for 640480 input image, the time needed is around 60-70ms
for 320240 input image, the time needed is around 20-30ms