Scribd
Upload
43 views3 pages

Assignment2 2223

The document outlines an assignment to design and implement a deep learning image classifier to recognize students' faces from close-up photos and register attendance. Students are asked to create a model that can classify at least 5 people with reasonable performance, and optionally implement additional features like handling more people or groups. The deliverables include source code, a report covering the model design and evaluation, and a demo video. The report, code, and video must be submitted by June 18th.

Uploaded by

Mr. Dhimal
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
43 views3 pages

Assignment2 2223

The document outlines an assignment to design and implement a deep learning image classifier to recognize students' faces from close-up photos and register attendance. Students are asked to create a model that can classify at least 5 people with reasonable performance, and optionally implement additional features like handling more people or groups. The deliverables include source code, a report covering the model design and evaluation, and a demo video. The report, code, and video must be submitted by June 18th.

Uploaded by

Mr. Dhimal
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

CSY3025 Artificial Intelligence Techniques - Assignment 2

Topic: Scan your face to register attendance

Submission deadline: 18th June 2023

Introduction

Attendance registration is an important tool for universities to improve student engagement


and performance. However, existing attendance tracking solutions have their limitations.

Your task is to design and implement a deep-learning image classifier that can recognise
students in the classroom from close-up facial images to register their attendance.

The classifier should have the capability of handling at least 5 (five) different people
including yourself with a reasonable performance. Acquire other students’ permission if you
wish to use photos of them for machine learning or testing. Other students’ identities
should be anonymised. You should discuss your choice of performance metrics according to
the application scenario. You are responsible to develop the image dataset for machine
learning and testing. You must also critically analyse any bias and ethical challenges related
to your dataset and model design. To achieve a higher grade, you are expected to design
and implement additional features.

Examples of additional features are (but are not limited to):


 Handling more people (>5)
 Handling multiple people in the same shot.
 Inclusion of different demographic groups,
 An exceptional performance demonstrated through “in-the-wild” testing
 implementations of user interfaces/user applications

To complete the task, you must study the problem space, establish a dataset, design and
train a deep learning single-label multi-class classifier, and use appropriate methods to
evaluate the performance of your classifier.

Deliverables

The deliverables of the assignment are:


 Source code of a working deep learning model implemented using Keras/TensorFlow
and Python. Source code of user applications, if applicable.
 A report that includes but is not limited to the following sections:
o Cover page
 A link to your dataset (e.g., github, figshare)
o Introduction
o Problem analysis and background research
o Building deep learning network (explain how you make choices at each
following step)
 Dataset
 DL Network (structure, loss function, optimiser, etc.)
 Training and evaluation
o Testing
o Summary of additional features (if applies)
o Discussions and conclusions
o References
o A copy of the source code in Appendix
 A 5-minute demo video about your design/implementation

There will be three submission points:


 Report
 Source code
 Demo video

You might also like