1. Roberto Rigamonti
  2. convolutional_filter_learning_matlab

Overview

HTTPS SSH
Convolutional filter learning - MATLAB version

This is a simplified implementation of the convolutional filter learning
algorithm presented in [1].
The code learns a filter bank on the images of the CIFAR-10 dataset
(http://www.cs.toronto.edu/~kriz/cifar.html).

The algorithm's settings are in the helper_functions/get_config.m file.
Before running the code, please make sure to have downloaded the MATLAB version
of the dataset and have properly filled the file containing the list of data
batches.


[1] R. Rigamonti, M. Brown, V. Lepetit, "Are Sparse Representations Really
    Relevant for Image Classification?", CVPR 2011