GPMS TRANSPORTAL APPLIED AI COURSES

• Here students work hands on in development of Computer Vision Solutions as selected by team - 5 Solutions in 5 Weeks, and 6th week for longer term project.
• Level - Intermediate - Choice - Specialisation (For Custom Solutions through industry partnerships).
• Prerequisites - Introduction to Computer Vision.
• Assessments - Micro Assessment weekly, Full Assessment last week.

---

Week 1 -Perfecting Object Detection

Learning Outcome

Students will perfect detection of any object through computer vision after understanding this with deep skill in creation detection for any use case.

1.1 Stack of Detection Algorithms with constraints Evaluation.
1.2 Medical and Precise Detection Use Case
1.3 Multi Object Use Case.

Practical Component
Practical Demonstrations on the entire stack working, use cases, and flaws, and then student created models demonstrated to students.

---

Week 2 - Multi Angle Face Recognition

Learning Outcome

Perfecting large data facial recognition and capture.

2.1 Integrating camera footage from various angles to detect person more accurately.
2.2 Addressing Similar Faces through Multi Model Approaches.
2.3 Lighting Conditions, and Data Set Amplification to address edge cases.

Practical Component
Practical Demonstrations on the entire stack working, use cases, and flaws, and then student created models demonstrated to students.

---

Week 3 - Integrating Depth and 3D Capture

Learning Outcome

Here the goal is to ensure algorithms predict the correct outcome independent of minor variations in inputs, and do not jump to alternate conclusions incorrectly.

3.1 Using depth based approaches with multi camera footage for capture of various metrics.
3.2 Understanding 3D coordinates of objects through multi cameras for different functionality creation.
3.3 Question Answering for various 3D Scenarios.

Practical Componenet
Practical Demonstrations the 3D objects, and arriving at outcomes from the input on contents thereof.

---

Week 4- Moving Drone Dataset Related Detection

Learning Outcome

Working with moving, and complex footage of drones, and multi sensor data.

4.1 Create drone based face detection, and activity detection.
4.2 Create drone based 3D capture through serial navigation footage.
4.3 Minimising model complexity for maximal processing on real time footage.

Practical Componenet
Practical Understanding of drone related moving footage, and capture mechanics.

---

Week 5- Integrating Multiple Models

Learning Outcome

Here the student should be able to use serial or parallel processing through multiple models to achieve accuracies for their task at Hand.

5.1 Transfer Learning and its working for specialised use cases.
5.2 Problem Break Up and Serial Model Approaches for Multi Class.
5.3 Integrating multi data set processing.

Practical Component
Practical Demonstrations of exact scenarios with outcomes for deeper understanding and hands on working.

---

Week 6-Project and Assessment

Learning Outcome

Here 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.

---