Topic:

You can choose any topic, such as face recognition, image recognition, covid-19 distribution and so on.

Requirements:

Must create a Create a computer vision application and deploy it to the edge-based inference platform.

Must have trained the model yourself

Must use a real data set (not fake data set, i.e. randomly generated numbers)

Must be original work (not a copy of a Kaggle project, etc.)

Two Page Report explaining your findings. Must be exactly two pages.

It’s best to have a five-minute speech draft

Colab Notebook Documentation:

Must explain in detail accuracy of model

Explain the tradeoffs you made

Do exploratory data exploration

There is detailed code and explanation in Colab.

The format will be as follows:

Required Format: Jupyter Notebook

Two Page Report explaining your findings. Must be exactly two pages.

You can leverage the GCP.

Project will be judged according to criteria

Reproducibility of the Project:Is every step reproducible in Jupyter notebook? Creativity:Did you create new insights that you shared to the class?

Craftsmanship:Did you take care in crafting your presentation and put thought into little details like thoughtful data visualization?

Difficulty: Was this a challenging problem you tackled?

ML Technique:How did you approach the problem?

Software Engineering Carpentry: Did you apply software engineering best practices? Resume Worthiness: Will this project get you a job?

How well does your two-page report explain your project?

You can get a general idea of the content according to the attachment.