基于贝叶斯框架融合的RGB-D图像显著性检测 |
| |
引用本文: | 王松涛,周真,靳薇,曲寒冰.基于贝叶斯框架融合的RGB-D图像显著性检测[J].自动化学报,2020,46(4):695-720. |
| |
作者姓名: | 王松涛 周真 靳薇 曲寒冰 |
| |
作者单位: | 1.哈尔滨理工大学测控技术与仪器省高校重点实验室 哈尔滨 150080 |
| |
基金项目: | 国家自然科学基金91746207北京市科技计划Z161100001116086 |
| |
摘 要: | 为了有效融合RGB图像颜色信息和Depth图像深度信息, 提出一种基于贝叶斯框架融合的RGB-D图像显著性检测方法.通过分析3D显著性在RGB图像和Depth图像分布的情况, 采用类条件互信息熵(Class-conditional mutual information, CMI)度量由深层卷积神经网络提取的颜色特征和深度特征的相关性, 依据贝叶斯定理得到RGB-D图像显著性后验概率.假设颜色特征和深度特征符合高斯分布, 基于DMNB (Discriminative mixed-membership naive Bayes)生成模型进行显著性检测建模, 其模型参数由变分最大期望算法进行估计.在RGB-D图像显著性检测公开数据集NLPR和NJU-DS2000上测试, 实验结果表明提出的方法具有更高的准确率和召回率.
|
关 键 词: | 贝叶斯融合 深度学习 生成模型 显著性检测 RGB-D图像 |
收稿时间: | 2017-05-02 |
Saliency Detection for RGB-D Images Under Bayesian Framework |
| |
Affiliation: | 1.The Higher Educational Key Laboratory for Measuring and Control Technology and Instrumentations of Heilongjiang Province, Harbin University of Science and Technology, Harbin 1500802.Research Center for Artificial Intelligence & Big Data Analysis, Beijing Academy of Science and Technology, Beijing 1000123.Key Laboratory of Big Data Analysis, Beijing Institute of New Technology Application, Beijing 100094 |
| |
Abstract: | In this paper, we propose a saliency detection model for RGB-D images based on the deep features of RGB images and depth images within a Bayesian framework. By analysis of 3D saliency in the case of RGB images and depth images, class-conditional mutual information (CMI) is computed for measuring the dependence of deep features extracted by CNN, then the posterior probability of the RGB-D saliency is formulated by applying the Bayes' theorem. By assuming that color- and depth-based deep features are Gaussian distributions, a discriminative mixed-membership naive Bayes (DMNB) model is used to calculate the final saliency map. The Gaussian distribution parameter can be estimated in the DMNB model by using a variational inference-based expectation maximization algorithm. The experimental results on the RGB-D image NLPR and NJU-DS2000 datasets show that the proposed model performs better than other existing models. |
| |
Keywords: | |
|
| 点击此处可从《自动化学报》浏览原始摘要信息 |
|
点击此处可从《自动化学报》下载全文 |
|