Decision Propagation Networks for Image Classification

TL;DR

This paper introduces Decision Propagation Networks that enhance image classification by propagating intermediate decisions from early to later layers, improving accuracy with minimal extra computation.

Contribution

It proposes a novel Decision Propagation Module (DPM) that explicitly encodes and refines decisions across network layers, advancing classification performance.

Findings

Significant accuracy improvements on four datasets.

Minimal additional computational cost.

Outperforms state-of-the-art methods.

Abstract

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.