Saliency Detection Combined with Small-Scale Deep Generation Model

CHEN Xing-yu; YE Feng; HUANG Tian-qiang; WENG Bin; CHEN Jia-zhen; LIN Wen-zhong

doi:10.12263/DZXB.20200488

您当前的位置：

首页 >

文章列表页 >

Saliency Detection Combined with Small-Scale Deep Generation Model

更新时间：2025-12-08

- Saliency Detection Combined with Small-Scale Deep Generation Model
- Acta Electronica Sinica Vol. 49, Issue 4, Pages: 768-774(2021)
- 作者机构：
  
  1. 福建师范大学数学与信息学院,福建,福州,350117
  2. 数字福建大数据安全技术研究所,福建,福州,350117
  3. 福建省信息处理与智能控制重点实验室(闽江学院),福建,福州,350108
  4. 福建师范大学数学与信息学院,福建,福州,350117
  5. 数字福建大数据安全技术研究所,福建,福州,350117
  6. 福建省信息处理与智能控制重点实验室(闽江学院),福建,福州,350108
- 作者简介：
- 基金信息：
- DOI：10.12263/DZXB.20200488
  CLC： TP391.41
- Published：2021
- 稿件说明：
移动端阅览
CHEN Xing-yu, YE Feng, HUANG Tian-qiang, et al. Saliency Detection Combined with Small-Scale Deep Generation Model[J]. Acta Electronica Sinica, 2021, 49(4): 768-774.
DOI：

CHEN Xing-yu, YE Feng, HUANG Tian-qiang, et al. Saliency Detection Combined with Small-Scale Deep Generation Model[J]. Acta Electronica Sinica, 2021, 49(4): 768-774. DOI： 10.12263/DZXB.20200488.

摘要

针对基于深度神经网络模型的显著性检测方法中存在的模型训练困难、模型参数量大以及检测速度慢等问题，本文提出了一种融合小型深度生成模型的显著性检测方法.方法以生成对抗网络为框架，设计了包含11个卷积模块和5个池化层的鉴别器网络以及不包含池化层，仅包含15个卷积模块和5个转置卷积模块的小型生成器网络.其中，小型生成器网络大小仅2.4M，参数量仅67万左右.将训练好的小型生成器用于显著性检测，并与LMB （融合背景块再选取过程的显著性检测）算法通过设计的融合算法进行融合，从而得到最终结果.通过大量的实验对比分析表明，提出的方法在

值和MAE （Mean Absolute Error）值上均取得大幅提升.

Abstract

Aiming at the difficulties of model training

large amount of model parameters

and slow detection speed in the saliency detection method based on deep neural network models

this paper proposes a saliency detection method that integrates small deep generative models. The method uses the generative adversarial network as the framework

designs a discriminator network consisting of 11 convolution modules and 5 pool layers

as well as a small generator network that does not contain the pool layer

including only 15 convolution modules and 5 transposed convolution modules. Among them

the size of the small generator network is only 2.4M

and the amount of parameters is only about 670

000. A trained small generator is used for

saliency detection and fused with the LMB (Salient object detection based on background block re-selection method) algorithm through the designed fusion algorithm to get the final result. A large number of experiments and comparative analysis show that the proposed method has achieved significant improvements in both

value and MAE (Mean Absolute Error) value.

关键词

Keywords

references

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Adversarial Learning and Enhanced Optimization Based Restoration Method for VC-Generated Speeches

Binary Code Similarity Detection Method Based on Cross-Modal Coordinated Representation Learning

FD-GAN: Frequency-Decomposed Generative Adversarial Network for Unpaired Underwater Image Enhancement

A Hierarchical Automatic Sleep Staging Model Based on Transformer Architecture

CR-GAN Complex Wireless Channel Modeling with Hidden Space Sampling and Hidden Feature Extraction

Related Author

SU Zhao-pin

ZHOU Xiao-lin

ZHANG Guo-fu

LIAN Chen-si

WANG Nian-song

YUE Feng

YANG Hong-yu

WANG Yun-long

Related Institution

Joint Laboratory of Intelligent Prevention and Recognition of Audio and Video

Intelligent Interconnected Systems Laboratory of Anhui Province (Hefei University of Technology)

Department of Physical Evidence Identification, Anhui Public Security Department

School of Computer and Information Technology, Hefei University of Technology

Key Laboratory of Civil Aviation Flight Networking, Civil Aviation University of China

⁰