Histogram Layers for Neural Engineered Features
This work addresses the challenge of combining engineered features with deep learning for computer vision practitioners, though it appears incremental as it adapts existing features rather than introducing a fundamentally new approach.
The paper tackles the problem of integrating traditional histogram-based features like local binary patterns into neural networks by proposing histogram layers that can learn these features, resulting in improved feature representation and image classification performance on benchmark and real-world datasets.
In the computer vision literature, many effective histogram-based features have been developed. These engineered features include local binary patterns and edge histogram descriptors among others and they have been shown to be informative features for a variety of computer vision tasks. In this paper, we explore whether these features can be learned through histogram layers embedded in a neural network and, therefore, be leveraged within deep learning frameworks. By using histogram features, local statistics of the feature maps from the convolution neural networks can be used to better represent the data. We present neural versions of local binary pattern and edge histogram descriptors that jointly improve the feature representation and perform image classification. Experiments are presented on benchmark and real-world datasets.