In computer vision, there are various different techniques to extract important features from images. These features are then later used in patten recognition, image classification, etc. Some of these techniques are comparable to some models of specific parts in human visual system.

In this project, we are interested in implementing some of the widely used techniques in feature learning (extraction) and applying them to a dataset of natural images. This usually corresponds to solving some optimization problem to find the features that represent the data more accurately.

The implementation can be either done in C or MATLAB (MATLAB is preferred).

And here are some lines to give you an idea about why we are interested in this project: Once the feature extraction techniques are implemented, the learned features will then be used as inputs to a neural network which mimics some parts of human memory (neural associative memory). The ultimate goal would be to see if one will get better information storage capacities in artificial neural memories when the inputs are natural stimuli (such as images) and pre-processed before being stored. HEre, pre-processing refers to the feature extraction procedure.

This project is suitable for students interested in computer vision, neural networks and mathematics who prefer a combination of theoretical and empirical works.

The prerequisites are: 1)Basic knowledge of linear algebra. 2)Being familiar with a suitable programming language (MATLAB is preferred. But C/C++ is acceptable.)

Some knowledge about feature learning models in computer vision is not necessary but is deeply encouraged.

The report is available via the following link:
Implementing Some Feature Extracting Techniques to Model Human Visual System