Deep Superpixel Cut for Unsupervised Image Segmentation

TL;DR

This paper introduces a novel deep unsupervised image segmentation method combining a Superpixelwise Autoencoder for feature learning and a Deep Superpixel Cut clustering algorithm, achieving effective segmentation without human annotations.

Contribution

The paper presents a new unsupervised segmentation framework that integrates deep autoencoding and a novel clustering algorithm, eliminating the need for labeled data.

Findings

Effective segmentation on BSDS500 dataset

Outperforms traditional unsupervised methods

Demonstrates strong deep feature learning capabilities

Abstract

Image segmentation, one of the most critical vision tasks, has been studied for many years. Most of the early algorithms are unsupervised methods, which use hand-crafted features to divide the image into many regions. Recently, owing to the great success of deep learning technology, CNNs based methods show superior performance in image segmentation. However, these methods rely on a large number of human annotations, which are expensive to collect. In this paper, we propose a deep unsupervised method for image segmentation, which contains the following two stages. First, a Superpixelwise Autoencoder (SuperAE) is designed to learn the deep embedding and reconstruct a smoothed image, then the smoothed image is passed to generate superpixels. Second, we present a novel clustering algorithm called Deep Superpixel Cut (DSC), which measures the deep similarity between superpixels and formulates image segmentation as a soft partitioning problem. Via backpropagation, DSC adaptively partitions the superpixels into perceptual regions. Experimental results on the BSDS500 dataset demonstrate the effectiveness of the proposed method.