Skip to content

Repository files navigation

Computer Vision 2024 Project [CSYS]

Product Classification & Verification Product classification and recognition is important task to automate tasks such as inventory management, pricing. For example, a retailer could use this technology to automatically identify and classify products in a store, and then use this information to update inventory levels, calculate products prices, and provide with information about the products. Project Objectives:

  1. Apply Image preprocessing or Feature Extraction techniques where needed.
  2. Train at least two classification model to determine the product category. You can use Classical computer vision or deep learning.
  3. Train a one/few shot learning model (e.g., Siamese) to verify if a product exists and recognize its type [Bonus].

Minimum Requirements:

a. You must use at least two classification model (classical or deep learning) for recognizing the product type.

b. You cannot use pixels directly as features in case of classical computer vision. You must use a feature extraction method first (e.g., BoW using SIFT) and then use a classifier like SVM or logistic regression.

Additional (Not required but bonus)

c. Train a Siamese model for one/few shot learning task.

About

Computer Vision 2024 FCIS project

Resources

Stars

0 stars

Watchers

2 watching

Forks

Releases

Packages

Used by

Contributors

Languages