Scribd Logo
Upload

Welcome to Scribd!

  • 168 views

    Computer Vision Nanodegree Syllabus: Before You Start

    Uploaded by

    Cristi Popa
    The Computer Vision Nanodegree program teaches students to become experts in computer vision through 3 courses over 3 months. Students will learn deep learning architectures and techniques like object classification, detection, segmentation, and how to apply these skills to applications such as image captioning, facial recognition, and landmark tracking. The program includes 3 projects where students build models to detect facial keypoints, generate image captions, and perform simultaneous localization and mapping. Students will gain skills in programming computer vision tasks and solving real-world problems with vision and deep learning.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Computer Vision Nanodegree Syllabus: Before You Start

    Uploaded by

    Cristi Popa
    168 views5 pages
    The Computer Vision Nanodegree program teaches students to become experts in computer vision through 3 courses over 3 months. Students will learn deep learning architectures and techniques like object classification, detection, segmentation, and how to apply these skills to applications such as image captioning, facial recognition, and landmark tracking. The program includes 3 projects where students build models to detect facial keypoints, generate image captions, and perform simultaneous localization and mapping. Students will gain skills in programming computer vision tasks and solving real-world problems with vision and deep learning.

    Original Description:

    Computer+Vision+Nanodegree+Syllabus

    Original Title

    Computer+Vision+Nanodegree+Syllabus

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The Computer Vision Nanodegree program teaches students to become experts in computer vision through 3 courses over 3 months. Students will learn deep learning architectures and techniques like object classification, detection, segmentation, and how to apply these skills to applications such as image captioning, facial recognition, and landmark tracking. The program includes 3 projects where students build models to detect facial keypoints, generate image captions, and perform simultaneous localization and mapping. Students will gain skills in programming computer vision tasks and solving real-world problems with vision and deep learning.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    168 views5 pages

    Computer Vision Nanodegree Syllabus: Before You Start

    Uploaded by

    Cristi Popa
    The Computer Vision Nanodegree program teaches students to become experts in computer vision through 3 courses over 3 months. Students will learn deep learning architectures and techniques like object classification, detection, segmentation, and how to apply these skills to applications such as image captioning, facial recognition, and landmark tracking. The program includes 3 projects where students build models to detect facial keypoints, generate image captions, and perform simultaneous localization and mapping. Students will gain skills in programming computer vision tasks and solving real-world problems with vision and deep learning.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    You are on page 1of 5

     

    Computer Vision Nanodegree 


    Syllabus   
    Become a Computer Vision Expert 
     
     
     
     
     
    Welcome to the Computer Vision Nanodegree program! 

    Before You Start 


     
    Educational Objectives​: In this program, you’ll learn the underlying math and programming concepts that 
    drive pattern recognition, object and image classification tasks, and object tracking systems. This course will 
    cover the latest in deep learning architectures used in industry, and you’ll combine current computer vision 
    and deep learning techniques to power a variety of applications. With the practical skills you gain in this 
    course, you’ll be able to program your own applications, extract information from any kind of image and 
    spatial data, and solve real-world challenges. 
     
    Prerequisite Knowledge​: In order to succeed in this program, we recommend having significant experience 
    with Python, and entry-level experience with probability and statistics, and deep learning architectures. 
    Specifically, we expect you to be able to write a class in Python and to add comments to your code for 
    others to read. Also, you should be familiar with the term “neural networks” and understand the differential 
    math that drives backpropagation. If you feel you need to add to your Python and statistics skills, we suggest 
    our M
    ​ achine Learning​ program. If you’d like to learn more about neural networks and backpropagation, 
    consider our D ​ eep Learning​ program. 
     
    Length of Program​: The program is comprised of 1 term, lasting 3 months. We expect students to work 10 
    hours/week on average. Make sure to set aside adequate time on your calendar for focused work.  
     
    Instructional Tools Available​: Video lectures, Jupyter notebooks, personalized project reviews.  

    Contact Info 
    While going through the program, if you have questions about anything, you can reach us at 
    [email protected].  
     

     
    version 1.0 
     

    Nanodegree Program Info 


    This program is designed to enhance your existing machine learning and deep learning skills with the 
    addition of computer vision theory and programming techniques. These computer vision skills can be 
    applied to various applications such as image and video processing, autonomous vehicle navigation, medical 
    diagnostics, smartphone apps, and much more. This program will not prepare you for a specific career or 
    role, rather, it will grow your deep learning and computer vision expertise, and give you the skills you need 
    to start applying computer vision techniques to real-world challenges and applications. 
     
    The term is comprised of 3 courses and 3 projects, which are described in detail below. Building a project is 
    one of the best ways to demonstrate the skills you've learned and each project will contribute to an 
    impressive professional portfolio that shows potential employers your mastery of computer vision and deep 
    learning techniques.  
     
     
    Length of Program​: 120 Hours​* 
    Number of Reviewed Projects​: 3 
     
    * The length of this program is an estimation of total hours the average student may take to complete all 
    required coursework, including lecture and project time. Actual hours may vary. 

       

     
     
    version 1.0
     

    Projects 
    Throughout this Nanodegree program, you'll master valuable skills by building the following projects: 
     
    ● Facial Keypoint Detection 
    ● Automatic Image Captioning 
    ● Landmark Detection and Tracking 
     
    In the sections below, you'll find a detailed description of each project along with the course material that 
    presents the skills required to complete the project. 

    Project: Facial Keypoint Detection 


    Use image processing techniques and deep learning techniques to detect faces in an image and find facial 
    keypoints, such as the position of the eyes, nose, and mouth on a face.  
     
    This project tests your knowledge of image processing and feature extraction techniques that allow you to 
    programmatically represent different facial features. You’ll also use your knowledge of deep learning 
    techniques to program a convolutional neural network to recognize facial keypoints. Facial keypoints include 
    points around the eyes, nose, and mouth on any face and are used in many applications, from facial 
    tracking to emotion recognition. 

    Introduction to Computer Vision 

    Lesson Title  Learning Outcomes 

    INTRODUCTION TO  ● Learn where computer vision techniques are used in industry. 
    COMPUTER VISION  ● Prepare for the course ahead with a detailed topic overview. 
    ● Start programming your own applications! 

    IMAGE REPRESENTATION  ● See how images are represented numerically. 


    AND ANALYSIS  ● Implement image processing techniques like color and 
    geometric transforms. 
    ● Program your own convolutional kernel for object 
    edge-detection. 

    CONVOLUTIONAL NN  ● Learn about the layers of a deep convolutional neural network: 
    LAYERS  convolutional, maxpooling, and fully-connected layers. 
    ● Build an CNN-based image classifier in PyTorch. 
    ● Learn about layer activation and feature visualization 
    techniques. 

    FEATURES AND OBJECT  ● Learn why distinguishing features are important in pattern and 
    RECOGNITION  object recognition tasks. 
    ● Write code to extract information about an object’s color and 

     
     
    version 1.0
     

    shape. 
    ● Use features to identify areas on a face and to recognize the 
    shape of a car or pedestrian on a road. 

    IMAGE SEGMENTATION  ● Implement k-means clustering to break an image up into parts. 


    ● Find the contours and edges of multiple objects in an image. 
    ● Learn about background subtraction for video. 
     

    Project: Automatic Image Captioning 


    Combine CNN and RNN knowledge to build a deep learning model that produces captions given an input 
    image. 
     
    Image captioning requires that you create a complex deep learning model with two components: a CNN that 
    transforms an input image into a set of features, and an RNN that turns those features into rich, descriptive 
    language. In this project, you will implement these cutting-edge deep learning architectures. 

    Advanced Computer Vision and Deep Learning 

    Lesson Title  Learning Outcomes 

    ADVANCED CNN  ● Learn about advances in CNN architectures. 


    ARCHITECTURES  ● See how region-based CNN’s, like Faster R-CNN, have allowed for 
    fast, localized object recognition in images. 
    ● Work with a YOLO/single shot object detection system. 

    RECURRENT NEURAL  ● Learn how recurrent neural networks learn from ordered 
    NETWORKS  sequences of data. 
    ● Implement an RNN for sequential text generation. 
    ● Explore how memory can be incorporated into a deep learning 
    model. 
    ● Understand where RNN’s are used in deep learning applications. 

    ATTENTION  ● Learn how attention allows models to focus on a specific piece 


    MECHANISMS  of input data. 
    ● Understand where attention is useful in natural language and 
    computer vision applications. 

    IMAGE CAPTIONING  ● Learn how to combine CNNs and RNNs to build a complex 
    captioning model. 
    ● Implement an LSTM for caption generation. 
    ● Train a model to predict captions and understand a visual scene. 

     
     
    version 1.0
     

    Project: Landmark Detection and Tracking 


    Use feature detection and keypoint descriptors to build a map of the environment with SLAM (simultaneous 
    localization and mapping). 
     
    Implement a robust method for tracking an object over time, using elements of probability, motion models, 
    and linear algebra. This project tests your knowledge of localization techniques that are widely used in 
    autonomous vehicle navigation. 

    Object Tracking and Localization 

    Lesson Title  Learning Outcomes 

    OBJECT MOTION AND  ● Learn how to programmatically track a single point over time. 
    TRACKING  ● Understand motion models that define object movement over 
    time.  
    ● Learn how to analyze videos as sequences of individual image 
    frames. 

    OPTICAL FLOW AND  ● Implement a method for tracking a set of unique features over 
    FEATURE MATCHING  time. 
    ● Learn how to match features from one image frame to another. 
    ● Track a moving car using optical flow. 

    ROBOT LOCALIZATION  ● Use Bayesian statistics to locate a robot in space. 


    ● Learn how sensor measurements can be used to safely navigate 
    an environment. 
    ● Understand Gaussian uncertainty. 
    ● Implement a histogram filter for robot localization in Python. 

    GRAPH SLAM  ● Identify landmarks and build up a map of an environment. 


    ● Learn how to simultaneously localize an autonomous vehicle 
    and create a map of landmarks. 
    ● Implement move and sense functions for a robotic vehicle. 
     
     
     

     
     
    version 1.0

    You might also like