Get updates on Eligibility, Admission, Placements Fees Structure
Quick Facts
| Medium Of Instructions | Mode Of Learning | Mode Of Delivery |
|---|---|---|
| English | Self Study | Video and Text Based |
Course Overview
The Modern Computer Vision certification course delves into classical and deep learning-based approaches to computer vision. Covering fundamental concepts like neurons and gradient descent, it discusses CNNs and traditional techniques such as edge detection and feature extraction. With industry support and expert guidance, students gain valuable skills for real-world applications.
The Modern Computer Vision certification by NPTEL equips students with the knowledge needed to tackle modern vision tasks. Students receive a well-rounded education from deep learning fundamentals to geometric vision techniques like 3D reconstruction. With support from various industry leaders, graduates are prepared for various career opportunities in research and industry.
The Highlights
Online Course
12 Weeks Duration
Offered by IIT Madras
Programme Offerings
- Certificate of completion
- Practical Learning
- Hands-on Experience
Courses and Certificate Fees
| Certificate Availability | Certificate Providing Authority |
|---|---|
| yes | IIT Madras (IITM) |
Eligibility Criteria
Academic Qualifications
While it is helpful to have some knowledge of image processing, linear algebra, and probability, it is not necessary for the Modern Computer Vision certification course.
Certification Qualifying Details
To receive the Modern Computer Vision certification by NPTEL, you need a minimum average assignment score of 10/25 and an exam score of 30/75.
What you will learn
After completing the Modern Computer Vision certification syllabus, you will delve into a comprehensive exploration of computer vision. You will learn fundamental concepts such as neurons, multilayer perceptron's, and gradient descent. You will also explore more advanced topics, including optimization, regularization, and convolutional neural networks (CNNs).
Upon completion of the Modern Computer Vision training, you will gain proficiency in traditional computer vision techniques, such as edge detection, feature extraction, and geometric vision tasks like camera geometry and 3D reconstruction. You will also learn how to apply these techniques to solve real-world vision problems effectively.
Who it is for
The Modern Computer Vision online course is suitable for individuals with a background in engineering or computer science looking to specialise in vision-related fields. This course can also be beneficial for:
Admission Details
Follow these steps to join the Modern Computer Vision classes:
Step 1: Browse the URL below:
https://swayam.gov.in/NPTEL
Step 2: Click on the “Sign-in/ Register” button
Step 3: Log in using a Google account-enabled email address
Step 4: Select your preferred course. Review the course details and fill out the required fields
Step 5: Click on the "Join" button
The Syllabus
Course introduction, Introduction to deep learning, Introduction to neuron
Multilayer perceptron (MLP), Gradient descent, Backpropagation in MLP
Optimization and regularization, Regularization and preprocessing, Convolutional neural network (CNN)
CNN properties, CNN architectures, Introduction to recurrent neural network (RNN), Encoder-Decoder models in RNN
Low-level vision, Spatial and frequency domain filtering, Edge detection
Line detection, Feature detectors, Harris corner detector
Blob detection, SIFT, Feature descriptors, SURF
Single-view geometry, 2D Geometric transformations, Camera intrinsic and extrinsic
Two-view stereo, Algebraic representation of epipolar geometry, Fundamental matrix computation
Structure from motion, Batch processing in SFM, Dense 3D reconstruction
Deep nets for stereo and SFM, Mid-level vision, Image segmentation
Deep nets for segmentation, High-level vision, Deep nets for object detection
Evaluation process
The Modern Computer Vision certification offers an optional exam. Your final score is determined by averaging the best 8 out of 12 assignments (25%) and your proctored exam score (75%).
IIT Madras (IITM) Frequently Asked Questions (FAQ's)
Familiarity with image processing, linear algebra, and probability is desirable but not mandatory. The course provides introductory materials to help students grasp essential concepts.
No prior experience with deep learning is required. The course starts with an introduction to deep learning concepts, making it accessible to beginners.
This course equips you with essential skills in both classical and deep learning-based approaches to computer vision, preparing you for various roles in research and industry.
Yes, the course incorporates practical assignments and projects to reinforce theoretical concepts. You will have the opportunity to apply what you have learned to solve practical computer vision problems.
Completing the course and its assignments makes you eligible for a certificate of completion from IIT Madras. This certification can serve as a valuable credential in your professional portfolio.