- Understanding the field, the process, and the mechanism of working of how the computer sees the world around
- Level - Basic - Mandatory - Introduction
- Prerequisites - None, No Coding knowledge necessary, will be built along with the course.
- Assessments - Micro Assessment weekly, Full Assessment last week.
Week 1 - What is Computer Vision
Learning OutcomeStudents should be able to understand, identify use cases of computer vision, and describe its underlying working
1.1 Introduction to Basic Classification (Dog, and Cat) use case.1.2 Extending the intelligence to multiple classes.
1.3 Extending to understanding faces, and objects.
Practical Component
Practical Demonstrations on working of various algorithms for the use case, to understand the need and innovation outcomes, working alongside to change parameters and understand key concepts and dependencies with the lab.
Week 2 - Computer Vision Intelligence Creation Overview
Learning OutcomeNow that we know the what, we focus on the how, what is the process of achieving the sort of intelligence across a Computer structure.
2.1 Primitive Methods, Overall Consideration, and Need for Ai Brain Technology.2.2 AI Brain Based Computer Vision and Its Working
2.3 Main Dependencies in Achieving sustainable intelligence (Data, Accuracy, Practical Considerations).
Practical Component
Understand the practical nuances of older algorithms through live simulations and understanding practical differences of various methods, and then solving problems of older methods using various AI Brain techniques. Understand how the results vary based on minor changes in important parameters.
Week 3 - Deep Dive Into Data Strategies
Learning OutcomeWhat is the role of Data, and how is it prepared and interpreted by the computer, across various data sources including colours, attributes, objects.
3.1 Basic Classification Data Preparation (Impact of incorrect preparation).3.2 Data Selection techniques and impact on intelligence outcome.
3.3 Identifying Data Needs for various Use Cases.
Practical Componenet
Practical Demonstrations showcasing the impact of Data Decisions on algorithm outcomes, and solving key issues related to the Data.
Week 4 - Deep Dive Into Algorithm Selection Strategies
Learning OutcomeHow to achieve effective training, and intelligence in minimal time first time accurate outcomes.
4.1 The current State of the Art for various tasks.4.2 Limitations and Accuracy Issues
4.3 Customization for your use case
Practical Componenet
Understanding of how algorithm selection impacts outcomes, and why some algorithms work better for some use cases.
Week 5-Deep Dive Into Assessment
Learning OutcomeHere the student should be able to completely assess and share a recommendation of the real world usability of various algorithms with practical trials, and outcomes envisioned.
5.1 Assessing correctness of outcomes, overfitting and various concepts.5.2 Practical failures of accurate models
5.3 Identifying cause of problem, and potential fix.
Practical Component
Demonstrations of algorithms that are overfit, ones that are trained incorrectly, and exercises to fix them within class, and identify causality.
Week 6-Project and Assessment
Learning OutcomeHere students work on a custom project of choice selected by the group to finish within a week from end to end with peer review, and complete presentation of outcomes.
6.1 Data Selection, Preparation, Algorithm Selection, and Initial Training Day.6.2 Initial Self Assessment, and Peer Assessment and review of outcomes.
6.3 Final Presentations and Scoring of outcome Achieved.
Practical Component
Practical Demonstrations of student made models at various stages, with fixes, and live demonstrations to a committee of peers or prospective recruiters.