Decision Propagation Networks for Image Classification
This work addresses a domain-specific problem in image classification by enhancing feature utilization, but it is incremental as it builds on existing network architectures.
The paper tackles the underexploitation of low-level features in early layers of convolutional neural networks for image classification by proposing a Decision Propagation Module (DPM) that extracts category-coherent guidance from early layers and propagates it to later layers, resulting in significant improvements on four datasets with minimal computational cost.
High-level (e.g., semantic) features encoded in the latter layers of convolutional neural networks are extensively exploited for image classification, leaving low-level (e.g., color) features in the early layers underexplored. In this paper, we propose a novel Decision Propagation Module (DPM) to make an intermediate decision that could act as category-coherent guidance extracted from early layers, and then propagate it to the latter layers. Therefore, by stacking a collection of DPMs into a classification network, the generated Decision Propagation Network is explicitly formulated as to progressively encode more discriminative features guided by the decision, and then refine the decision based on the new generated features layer by layer. Comprehensive results on four publicly available datasets validate DPM could bring significant improvements for existing classification networks with minimal additional computational cost and is superior to the state-of-the-art methods.