基于贝叶斯框架融合的RGB-D图像显著性检测

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.

Saliency Detection for RGB-D Images Under Bayesian Framework

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.