Convolutional Neural Networks from Image Markers
This work provides an alternative, potentially more efficient, method for training convolutional neural networks for researchers and practitioners in image classification, especially when dealing with limited data or computational resources.
The paper extends the Feature Learning from Image Markers (FLIM) technique to include fully connected layers, enabling the estimation of convolutional filters without backpropagation. The extended FLIM-based convolutional neural networks are shown to outperform architectures trained from scratch using backpropagation on various image classification problems.
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.