Contour Detection Using Cost-Sensitive Convolutional Neural Networks

TL;DR

This paper presents a novel contour detection method using cost-sensitive DenseNet CNNs combined with SVM classification, achieving competitive results on the BSDS500 dataset.

Contribution

It introduces a cost-sensitive fine-tuning approach for DenseNet to improve per-pixel feature extraction for contour detection.

Findings

Achieves comparable performance to state-of-the-art methods on BSDS500.

Demonstrates effectiveness of multi-layer CNN features for contour detection.

Shows benefits of cost-sensitive learning with small datasets.

Abstract

We address the problem of contour detection via per-pixel classifications of edge point. To facilitate the process, the proposed approach leverages with DenseNet, an efficient implementation of multiscale convolutional neural networks (CNNs), to extract an informative feature vector for each pixel and uses an SVM classifier to accomplish contour detection. The main challenge lies in adapting a pre-trained per-image CNN model for yielding per-pixel image features. We propose to base on the DenseNet architecture to achieve pixelwise fine-tuning and then consider a cost-sensitive strategy to further improve the learning with a small dataset of edge and non-edge image patches. In the experiment of contour detection, we look into the effectiveness of combining per-pixel features from different CNN layers and obtain comparable performances to the state-of-the-art on BSDS500.