TrainYourOwnYOLO: Building a Custom Object Detector from Scratch

This repo let's you train a custom image detector using the state-of-the-art YOLOv3 computer vision algorithm. For a short write up check out this medium post.

Pipeline Overview

To build and test your object detection algorithm follow the below steps:

1. Image Annotation
   • Install Microsoft's Visual Object Tagging Tool (VoTT)
   • Annotate images
2. Training
   • Download pre-trained weights
   • Train your custom YOLO model on annotated images
3. Inference
   • Detect objects in new images and videos

Repo structure

• 1_Image_Annotation: Scripts and instructions on annotating images
• 2_Training: Scripts and instructions on training your YOLOv3 model
• 3_Inference: Scripts and instructions on testing your trained YOLO model on new images and videos
• Data: Input Data, Output Data, Model Weights and Results
• Utils: Utility scripts used by main scripts

Getting Started

Requisites

The only hard requirement is a running version of python 3.3 or newer. To install the latest python 3.x version go to

• python.org/downloads

and follow the installation instructions.

To speed up training, it is recommended to use a GPU with CUDA support. For example on AWS you can use a p2.xlarge instance (Tesla K80 GPU with 12GB memory). Inference is very fast even on a CPU with approximately ~2 images per second.

Installation

Setting up Virtual Environment [Linux or Mac]

Clone this repo with:

git clone https://github.com/AntonMu/TrainYourOwnYOLO
cd TrainYourOwnYOLO/

Create Virtual (Linux/Mac) Environment (requires venv which is included in the standard library of Python 3.3 or newer):

python3 -m venv env
source env/bin/activate

Make sure that, from now on, you run all commands from within your virtual environment.

Setting up Virtual Environment [Windows]

Use the Github Desktop GUI to clone this repo to your local machine. Navigate to the TrainYourOwnYOLO project folder and open a power shell window by pressing Shift + Right Click and selecting Open PowerShell window here in the drop-down menu.

Create Virtual (Windows) Environment (requires venv which is included in the standard library of Python 3.3 or newer):

py -m venv env
.\env\Scripts\activate

Make sure that, from now on, you run all commands from within your virtual environment.

Install Required Packages [Windows, Mac or Linux]

If you are running TrainYourOwnYOLO on a machine with GPU with CUDA drivers installed run:

pip3 install -r requirements_gpu.txt

Otherwise, run:

pip3 install -r requirements_cpu.txt

Quick Start (Inference only)

To test the cat face detector on test images located in TrainYourOwnYOLO/Data/Source_Images/Test_Images run the Minimal_Example.py script in the root folder with:

python Minimal_Example.py

The outputs are saved in TrainYourOwnYOLO/Data/Source_Images/Test_Image_Detection_Results. This includes:

• Cat pictures with bounding boxes around faces with confidence scores and
• Detection_Results.csv file with file names and locations of bounding boxes.

If you want to detect cat faces in your own pictures, replace the cat images in Data/Source_Images/Test_Images with your own images.

Full Start (Training and Inference)

To train your own custom YOLO object detector please follow the instructions detailed in the three numbered subfolders of this repo:

• 1_Image_Annotation,
• 2_Training and
• 3_Inference.

To make everything run smoothly it is highly recommended to keep the original folder structure of this repo!

Each *.py script has various command line options that help tweak performance and change things such as input and output directories. All scripts are initialized with good default values that help accomplish all tasks as long as the original folder structure is preserved. To learn more about available command line options of a python script <script_name.py> run:

python <script_name.py> -h

License

Unless explicitly stated otherwise at the top of a file, all code is licensed under the MIT license. This repo makes use of ilmonteux/logohunter which itself is inspired by qqwweee/keras-yolo3.

Acknowledgements

Many thanks to Niklas Wilson for contributing towards making this repo compatible with Tensorflow 2.0.

Troubleshooting

0. If you encounter any error, please make sure you follow the instructions exactly (word by word). Once you are familiar with the code, you're welcome to modify it as needed but in order to minimize error, I encourage you to not deviate from the instructions above.

1. If you are using pipenv and are having trouble running python3 -m venv env, try:

   pipenv shell
   

2. If you are having trouble getting cv2 to run, try:

   apt-get update
   apt-get install -y libsm6 libxext6 libxrender-dev
   pip install opencv-python
   

3. If you are a Linux user and having trouble installing *.snap package files try:

   snap install --dangerous vott-2.1.0-linux.snap
   

   See Snap Tutorial for more information.

Stay Up-to-Date

• ⭐ star this repo to get notifications on future improvements and
• 🍴 fork this repo if you like to use it as part of your own project.