SETUP – PART V

Today, you will be learning, how to do Object Detection with your dataset using YOLOv3 model

Previous parts contain the theory and basics of YOLOv3 and Object Detection, blah blah blah. But the upcoming blog series will be purely practical. Better do it while you read for pure understanding.

Steps to be followed

  • Download the dataset and fix it
  • I will provide the link for the tool to annotate your dataset
  • Using labelImg, Annotate your dataset in YOLO format
  • Thanks to Google Colab for giving us free GPU and a notebook to run our Deep Learning models – A link for the code is attached
  • Run the model and Test your model finally!
  • Let us get started

LabelImg

1. Prepare the Image dataset

An image dataset is a folder contains a lot of images (I suggest getting at least 100 of them) where there is the custom object you want to detect. For example I am training YOLO to recognize a Panda, so I have downloaded around 150 images containing Panda images.

I have collected Panda images and saved them in a directory named images

https://github.com/tzutalin/labelImg -> Link to download and clone the labelImg repository, make sure that you follow the instructions to set up and install labelImg with necessary dependencies

Open labelImg, load the directory

Annotate the images for what object you are going to detect

  1. Once we run LabelImg let’s click on “Open Dir”.
  2. We choose the folder where the images are located
  3. Then we click on – Select folder
  4. We then click on – Change save dir
  5. We select the folder where the images are located (same folder we selected on step 2)
  6. Then we click on – Select folder
  7. Finally make sure that we’re using the settings for YOLO
    If pascalVOC is written, then let’s click and we will see YOLO
Click pascalVOC, it’ll change to Yolo

 2. Now we’re ready to label the images.

  1. Let’s click on – Create RectBox
  2. Let’s select the area where our object is located (in my case I’m going to select the Panda)
  3. We add the label with the name of our object. In my case typed Panda and press Ok.
  4. We click on “Save”

We are going to do this operation for all the images we have on the dataset.

You will be seeing the training and testing in the Next part

Yo Panda🐼- I’m coming for you🧔

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s