حول هذه الدورة
دورة Deep Learning for Computer Vision هي دورة شاملة تبدأ من الأساسيات وصولاً إلى المواضيع المتقدمة في Computer Vision باستخدام Deep Learning. سيتعرف المتعلمون في هذه الدورة على كيفية استخدام Convolutional Neural Networks (CNNs) في تطبيقات مثل object detection و image classification، كما سيتعلمون تقنيات Transfer Learning لتحسين الأداء باستخدام pre-trained models. الدورة تشمل أيضًا تقنيات متقدمة في video analysis و motion tracking.
أهداف التعلم
المواد
- Deep Learning (e.g., artificial neural networks).
- Convolutional Neural Networks (CNNs) (architecture of CNNs, convolutional layers,
- pooling layers, activation functions, loss functions, training and optimization, etc.)
- Transfer Learning (e.g., using pre-trained models, fine-tuning, transfer learning in CNNs,etc.)
- Object Detection and Recognition
- Video Analysis and Tracking
- Optical Flow (e.g., motion estimation, Lucas-Kanade method, Horn-Schunck method, etc.)
- Motion Estimation (e.g., background subtraction, optical flow, feature tracking, etc.)
- Tracking and Surveillance (e.g., object tracking, multi-object tracking, Kalman filter, particle filter, etc.)
- Bose Estimation
متطلبات
- معرفة أساسية بـ Python.
- أساسيات الرياضيات: معرفة بالجبر الخطي والإحصاء ستكون مفيدة.
- أساسيات التعلم الآلي (Machine Learning) ستكون مفيدة، لكن ليست ضرورية.
الجمهور المستهدف
- طلاب علوم الكمبيوتر وعلوم البيانات الذين يرغبون في تعلم Deep Learning و Computer Vision.
- المطورون الذين يريدون اكتساب مهارات في object detection و image classification.
- المهنيون في الذكاء الاصطناعي الذين يودون تحسين مهاراتهم في Computer Vision باستخدام deep learning.
- المهتمون بتطبيق تقنيات الرؤية الحاسوبية في الصناعات المختلفة مثل الطب، الأمن، و السيارات الذاتية القيادة.
منهاج دراسي
Deep Learning
Deep Learning Fundamentals: From History to Single-Layer Perceptrons03:15:18
Quiz1
Building Deep Neural Networks: Activation Functions and Keras Hands-On02:35:58
Mastering TensorFlow: An Introduction and Practical Guide02:58:55
Quiz2
Training Neural Networks: Backpropagation, Loss Functions, and Real-World Practice02:38:45
Quiz3
Optimization in Machine Learning: Bias-Variance, Overfitting, and Error Analysis02:57:14
Quiz4
Why CNNs Outperform MLPs for Images: Components and Basics3:11:53
Quiz5
Advanced CNN Techniques: Image Augmentation, Transfer Learning, and MNIST Practice02:38:31
Quiz6
Object Detection Fundamentals: R-CNN Explained01:24:45
Fast R-CNN and YOLO: Object Detection Simplified01:43:24
YOLO Recap and Introduction to Object Tracking03:34:18
Hands-On Object Detection: Practical Applications03:00:18
Quiz7
Object Detection and Tracking: Practical Deep SORT in Action02:28:52
Quiz8
3D Face Reconstruction, Facial Keypoint Detection, and Transfer Learning02:47:24
Quiz9
Solving the OCR Problem: An Introduction to Optical Character Recognition03:13:06
Quiz1O
احصل على شهادة
أضف هذه الشهادة إلى سيرتك الذاتية لإثبات مهاراتك وزيادة فرصك في لفت الانتباه.
أساتذتك
Mohamed ElDakdouky
Senior Computer Vision Enginer at ServiceNow
As a computer vision engineer with over two years of experience, I am passionate about developing cutting-edge computer vision solutions that drive innovation and solve complex problems for a range of industries. My areas of expertise include image processing, object detection, deep learning, and machine vision, which allow me to design and implement efficient, reliable, and scalable computer vision systems using popular frameworks such as TensorFlow, PyTorch, and OpenCV. Additionally, I am skilled in utilizing advanced tools such as DeepStream, TensorRT, and OpenVINO, as well as programming languages such as C++, to accelerate deep learning inference, optimize deep learning models, and deploy them on edge devices. In my previous roles, I have demonstrated my ability to collaborate effectively with cross-functional teams and stakeholders to identify business objectives and deliver solutions that exceed expectations. Specifically, I have experience in developing computer vision applications for surveillance, autonomous vehicles, and robotics, as well as customizing off-the-shelf algorithms for specific use cases. I am committed to staying up-to-date with the latest trends and techniques in computer vision and constantly exploring new tools to enhance my skills
المواد
- Deep Learning (e.g., artificial neural networks).
- Convolutional Neural Networks (CNNs) (architecture of CNNs, convolutional layers,
- pooling layers, activation functions, loss functions, training and optimization, etc.)
- Transfer Learning (e.g., using pre-trained models, fine-tuning, transfer learning in CNNs,etc.)
- Object Detection and Recognition
- Video Analysis and Tracking
- Optical Flow (e.g., motion estimation, Lucas-Kanade method, Horn-Schunck method, etc.)
- Motion Estimation (e.g., background subtraction, optical flow, feature tracking, etc.)
- Tracking and Surveillance (e.g., object tracking, multi-object tracking, Kalman filter, particle filter, etc.)
- Bose Estimation
