Introduction to Yolo V3 Object Detection - FULL COURSE

Introduction to Yolo V3 Object Detection

⭐6-in-1 AI MEGA Course - https://augmentedstartups.info/AugmentedAICVPRO ► Yolo V3 Coupon: http://augmentedstartups.info/YoloV3_Course ► Github - http://augmentedstartups.info/github-yolo-access

Welcome to the Practical computer vision course on how to implement YoloV3 Objection Detection, using Deep Learning. The only course that teaches you how to develop real-time AI object detection, using the state-of-art CNN model called Yolo version 3. You will learn the full practical workflow from executing all the way to training your models.

Qualifications So who am I to teach you, well my Name is Ritesh Kanjee and I have a Masters Degree in Electronic engineering majoring in computer vision and AI. I have over 43000 students on Udemy and 63k subscribers on my YouTube channel Augmented Startups, teaching people from 147 countries around the world.

How will I get it?

Because we know that you are very busy and you just want to get something up and running, we don’t want to waste your time with a whole lot of theory that is already freely available on the net. That is why this course is a practical approach to helping you get YoloV3 up and running.

So, we start off this course and dive right in with the swift execution of a pretrained YoloV3 model on both recorded video as well as with a live stream from a webcam.

Section 3 is focused on training as well the yolo v3 workflow that uses Supervisely to accelerate our training process.

For this, we first setup our deep learning cluster then we search the web for all sorts of images for our dataset. Once we have those images we need to annotate or label them. If you have tons of images and don’t want to label them, then I will introduce you to Human in the Loop Annotation, which is essentially an AI Semi-autonomous annotation method. I show you how this works in this lecture.

The next magic trick I have under my sleeve is Data augmentation. Image or Data augmentation artificially creates training images through different ways of processing or combination of multiple processing, such as random rotation, shifts, shear and flips, etc. Data augmentation basically helps us to improve our model’s performance.

Once our data has been augmented, we are now ready to train our network.

Training produces the checkpoint weights which we can use in our original yolo v3 object detector but rather for our own custom computer vision applications.

In Section 4 we dive deeper into the code for a better understanding of the OpenCV functions used.

And then finally we conclude the course with external resources as well as some bonus tips.

Who is this course for? This course is for those who have some experience in computer vision or deep learning and who want ... https://www.youtube.com/watch?v=hUWufu2EkXk