RED-NET: A Recursive Encoder-Decoder Network for Edge Detection

TL;DR

RED-NET introduces a recursive encoder-decoder architecture with skip-connections that enhances edge detection performance by increasing network depth efficiently and preserving fine details, achieving state-of-the-art results.

Contribution

The paper presents a novel recursive encoder-decoder network with skip-connections for edge detection, improving depth and detail preservation without increasing parameters.

Findings

Achieves superior edge detection metrics on BSDS500, NYUD, and Pascal Context datasets.

Significantly outperforms previous methods on standard evaluation metrics.

Demonstrates the effectiveness of recursive and skip-connection design in deep edge detection networks.

Abstract

In this paper, we introduce RED-NET: A Recursive Encoder-Decoder Network with Skip-Connections for edge detection in natural images. The proposed network is a novel integration of a Recursive Neural Network with an Encoder-Decoder architecture. The recursive network enables us to increase the network depth without increasing the number of parameters. Adding skip-connections between encoder and decoder helps the gradients reach all the layers of a network more easily and allows information related to finer details in the early stage of the encoder to be fully utilized in the decoder. Based on our extensive experiments on popular boundary detection datasets including BSDS500 \cite{Arbelaez2011}, NYUD \cite{Silberman2012} and Pascal Context \cite{Mottaghi2014}, RED-NET significantly advances the state-of-the-art on edge detection regarding standard evaluation metrics such as Optimal Dataset Scale (ODS) F-measure, Optimal Image Scale (OIS) F-measure, and Average Precision (AP).