In this paper, a deep network based on Local Binary Pattern (LBP) descriptor is proposed, which is named as Local Binary Pattern Network (LBPNet). Two filters are used in LBPNet, which are based on LBP and Principle Component Analysis (PCA) techniques, respectively. The over-complete patch-based features are extracted hierarchically using these two filters. After feature extraction, the LBPNet employs a simple network to measure the similarity of the extracted features. Major characteristics of the proposed LBPNet are summarized in the following:
Feature extraction in dense grid: Both of the two filters are replicated densely in layers.
Multi layer architecture: The representations are extracted hierarchically: the latter layer extracts a higher level of abstractions from the lower ones of the earlier layer.
Partially connected layer: Filters only compute based on the selected subset of the inputs from the earlier layer.
Multi-scale analysis: Filters with different parameters are used in each of the layers to capture multi-scale statistics.