Convolutional Neural Networks from Image Markers

TL;DR

This paper extends the FLIM technique to fully connected layers, enabling CNNs to learn filters from minimal user-drawn markers without backpropagation, and demonstrates superior performance over traditional training methods.

Contribution

The paper introduces an extension of FLIM for fully connected layers and evaluates its effectiveness across various image classification tasks.

Findings

FLIM-based CNNs outperform traditional trained CNNs.

Marker selection impacts classification accuracy.

Adding fully connected layers enhances model performance.

Abstract

A technique named Feature Learning from Image Markers (FLIM) was recently proposed to estimate convolutional filters, with no backpropagation, from strokes drawn by a user on very few images (e.g., 1-3) per class, and demonstrated for coconut-tree image classification. This paper extends FLIM for fully connected layers and demonstrates it on different image classification problems. The work evaluates marker selection from multiple users and the impact of adding a fully connected layer. The results show that FLIM-based convolutional neural networks can outperform the same architecture trained from scratch by backpropagation.