Image classification: cats vs. dogs

Oge Marques, Florida Atlantic University, EECS
Author Profile

Summary

In this project you will design and implement a deep learning solution for image classification based on images of cats and dogs.

Share your modifications and improvements to this activity through the Community Contribution Tool »

Learning Goals

Learning objectives:

- Learn how to implement a complete, fully functional, deep learning solution for a contemporary computer vision challenge using MATLAB.
- Get acquainted with representative contemporary datasets, challenges and problems in computer vision and machine learning.
- Learn how to use deep neural networks under the paradigm of transfer learning.
- Demonstrate the ability to perform model selection, fine-tuning, and performance evaluation of different solutions to the same problem.

Context for Use

This is a lab project for an upper undergraduate or early graduate-level course in Digital Image Processing, Computer Vision, or Deep Learning. It could be used as the final (term) project for the course.

Students will typically need 2-3 weeks to produce high-quality results. They should be familiar with MATLAB, Image Processing, Machine Learning and Deep Learning (at least at the level of the corresponding Onramps).

Description and Teaching Materials

(See attached PDF)
Image classification using deep learning (Acrobat (PDF) 399kB Sep10 21)
Cats vs. dogs: starter code (Matlab File 11kB Sep10 21)


Teaching Notes and Tips


Assessment

A typical rubric for this type of assignment would be as follows:
- Code (correctness): 40 %
- Code (style and documentation): 20 %
- Report (contents, format, language): 40 %

References and Resources