Projects
Assignments
Overview
This course provides a comprehensive understanding of deep learning techniques for autonomous driving. Participants will gain practical experience in developing, training, and deploying models using artificial neural networks (ANN) and convolutional neural networks (CNN). The course also covers object detection using YOLO, image processing with OpenCV, and deploying models on embedded platforms like Jetson Nano. Participants will learn to solve real-world challenges in autonomous driving using advanced deep learning applications.
Objectives
By the end of this course, leaner will be able to:
-
Understand fundamental machine learning concepts and their applications in autonomous driving
-
Develop and train ANN and CNN models for image classification and object detection
-
Apply OpenCV for image processing and computer vision tasks
-
Implement object detection using YOLO for real-time applications
-
Deploy and optimize deep learning models on embedded platforms like Jetson Nano
-
Evaluate and enhance the performance of deployed models
Prerequisites
-
Python Basics
-
Machine Learning Basics
Course Outline
-
Deep learning fundamentals and applications in autonomous driving
-
Key perception tasks: object detection, classification, segmentation, depth estimation
-
Overview of deep learning frameworks
-
Introduction to Jetson Nano and setting up the development environment
-
Python libraries for deep learning (NumPy, Pandas)
-
ANN architecture: neurons, layers, and activation functions
-
Perceptron and multilayer perceptron (MLP) models
-
Backpropagation and gradient descent optimization
-
Developing and training ANN models using TensorFlow/Keras
-
Hands-on: Implementing ANNs for classification tasks
-
Introduction to CNN architecture: convolutional layers, pooling layers, fully connected layers
-
Understanding feature extraction and representation learning
-
Training CNN models with TensorFlow/Keras
-
Transfer learning and fine-tuning pre-trained models
-
Hands-on: Implementing CNNs for image classification and object detection
-
Basics of OpenCV for image and video processing
-
Image manipulation, filtering, and feature extraction
-
Object detection and tracking using traditional computer vision techniques
-
Integrating OpenCV with deep learning models
-
Hands-on: Implementing image processing pipelines using OpenCV
-
Introduction to YOLO (You Only Look Once) and its versions (YOLOv3, YOLOv4, YOLOv5)
-
Training and evaluating YOLO models for custom datasets
-
Model optimization for enhanced performance and accuracy
-
Hands-on: Implementing object detection using YOLO on Jetson Nano
-
Deploying deep learning models on Jetson Nano
-
Model optimization techniques including quantization and pruning
-
Real-time performance considerations and integration with autonomous driving systems
-
Hands-on: Deploying and optimizing YOLO for real-time object detection
