Advanced Computer Vision-1
In this training program, you will learn to recognize image or video data using fundamental Computer Vision methods and algorithms, and to build models based on images to achieve useful results. In a world evolving through the possibilities created by Machine Learning, solutions like Midjourney, Stable Diffusion, SAM, DALL-E, and others now allow you to create any image you can imagine and develop various solutions. This training program will teach you advanced concepts, cutting-edge techniques, and practical applications in the field of Computer Vision. After the training, you will possess all the technical knowledge required to either create your own startup or work as a professional Computer Vision expert in any company. This training will serve as an ideal starting point for joining the ranks of researchers contributing to the magical world of Machine Learning.
Training table
Information about the training
Computer Vision algorithms are widely used in agriculture for recognizing diseases from plant images, in medicine for image-based diagnostics, in space exploration, security, and surveillance, and are among the new artificial intelligence solutions that reduce costs. In a world where Artificial Intelligence and Machine Learning are rapidly expanding, an example of the use of Computer Vision algorithms is in self-driving cars, where models can issue warnings. The recognition of humans, animals, and other objects based on images by computers is widely applied in many industries and government agencies to increase efficiency and help prevent fraud. In the Advanced Computer Vision - 1 training program, you will learn through real projects to apply face recognition systems, segmentation, advanced CNN architectures, and solve problems involving sequences in images using RNNs. Through Python, you will gain the skills to apply Computer Vision algorithms using libraries such as OpenCV, numpy, Pytorch, Keras, and TensorFlow. During the training, you will work on practical computer vision solutions such as face recognition systems, brain organ segmentation and detection from MRI images, and text recognition from images.
Who is this training for?
Students
Programmers
Data Science Experts
Artificial Intelligence Enthusiasts
Researchers and Academics
Certificate
It provides a certificate to those who complete the training as Certified Data Scientist and others. You can see a sample certificate on the right.
How Computer Vision Works
How Computer Vision Works
Google Cloud Tech
In this video we'll uncover the magic of computer vision models by breaking down how Convolutional Neural Nets work under the hood, and we'll end with a live demo of the Vision API
Syllabus
Session 1
Introduction to Face Recognition Systems and Segmentation
Face Recognition:
Introduction to Face Recognition Systems
Face Similarity with Keras VGGFace
One-Shot Learning with Keras and Face Recognition with PyTorch FaceNet
DeepFace for Age, Gender, Emotion, Nationality, and Face Recognition
Segmentation Introduction:
Contours: Drawing, Hierarchy, and Modes
Moments, Sorting, Approximating, and Matching Contours
Line and Circle Detection
Counting Contours and Ellipses, Template Matching
Corner Detection
Case Study 1
Building a Face Recognition System and Logging Data to Database
Session 2
Introduction to Image Segmentation and Popular Algorithms
Details of FCN Architecture
Upsampling Methods
Encoder and Decoder
Measuring Model Performance with IoU (Intersection over Union) and Dice Score
U-Net Model, U-Net Encoder and Decoder
Case Study 2
Segmentation of Corpus Callosum from Brain MRI Images using U-Net
Session 3
R-CNN (Region-based Convolutional Neural Networks) for Image Recognition
Object Detection
Non-Maximum Suppression
R-CNNs, Fast R-CNNs, and Faster R-CNNs
Single Shot Detectors (SSDs)
Introduction to YOLO
Training a Model with YOLO
EfficientDet
Detectron2
Case Study 3
Using YOLO to Detect Corpus Callosum in Brain MRI Images
Session 4
Introduction to RNNs
Simple RNN / Elman Unit
RNN Code Preparation
GRU and LSTM
Video Classification with RNN and CNN
Case Study 4
Reading Text from Images using CRNN (Convolutional Recurrent Neural Network)