.png){kind=icon}
返回
DIC: Deep Image Clustering for Unsupervised Image Segmentation
.png){kind=icon}
OA
AI
DOI：10.1109/ACCESS.2020.2974496.png)
.png){kind=icon}
.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Unsupervised segmentation is an essential pre-processing technique in many computer vision tasks. However, current unsupervised segmentation techniques are sensitive to the parameters such as the segmentation numbers or of high training and inference complexity. Encouraged by neural networks flexibility and their ability for modelling intricate patterns, an unsupervised segmentation framework based on a novel deep image clustering (DIC) model is proposed. The DIC consists of a feature transformation subnetwork (FTS) and a trainable deep clustering subnetwork (DCS) for unsupervised image clustering. FTS is built on a simple and capable network architecture. DCS can assign pixels with different cluster numbers by updating cluster associations and cluster centers iteratively. Moreover, a superpixel guided iterative refinement loss is designed to optimize the DIC parameters in an overfitting manner. Extensive experiments have been conducted on the Berkley Segmentation Database. The experimental results show that DCS is more effective in aggregating features during the clustering procedure. DIC has also proven to be less sensitive to varying segmentation parameters and of lower computation costs, and DIC can achieve significantly better segmentation performance compared to the state-of-the-art techniques. The source code is available on ttps://github.com/zmbhou/DIC.
Keyword:
Image segmentation
Convolution
Neural networks
Training
Feature extraction
Network architecture
Synthetic aperture sonar
Unsupervised segmentation
deep image clustering
deep clustering subnetwork
iterative refinement loss
overfitting training
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
3.6
论文数:
9.8W
被引数:
29.4W
机构
暂无机构信息
引用论文
Tinnitus Retraining Therapy (TRT) as a Method for Treatment of Tinnitus and Hyperacusis Patients耳鸣再训练疗法 (TRT) 作为治疗耳鸣和高亢患者的方法
PREAI
分享收藏Superpixel based continuous conditional random field neural network for semantic segmentation.png)
NEUROCOMPUTING
NEUROCOMPUTINGIF6.5
PREAI
分享收藏分享收藏
查看更多内容