Distributed Iterative Gating Networks for Semantic Segmentation

TL;DR

This paper introduces DIGNet, a novel neural network structure with cascaded feedback for improved semantic segmentation, demonstrating superior performance over traditional feed-forward and recurrent models on multiple datasets.

Contribution

The paper proposes a lightweight, feedback-based neural architecture with cascaded feedback mechanisms for enhanced information flow in semantic segmentation.

Findings

DIGNet outperforms baseline models on PASCAL VOC 2012, COCO-Stuff, and ADE20K datasets.

Cascaded feedback in DIGNet is more effective than other feedback strategies.

Recurrent feedback improves pixel-wise labeling accuracy.

Abstract

In this paper, we present a canonical structure for controlling information flow in neural networks with an efficient feedback routing mechanism based on a strategy of Distributed Iterative Gating (DIGNet). The structure of this mechanism derives from a strong conceptual foundation and presents a light-weight mechanism for adaptive control of computation similar to recurrent convolutional neural networks by integrating feedback signals with a feed-forward architecture. In contrast to other RNN formulations, DIGNet generates feedback signals in a cascaded manner that implicitly carries information from all the layers above. This cascaded feedback propagation by means of the propagator gates is found to be more effective compared to other feedback mechanisms that use feedback from the output of either the corresponding stage or from the previous stage. Experiments reveal the high degree of capability that this recurrent approach with cascaded feedback presents over feed-forward baselines and other recurrent models for pixel-wise labeling problems on three challenging datasets, PASCAL VOC 2012, COCO-Stuff, and ADE20K.