WavNet — Visual saliency detection using Discrete Wavelet Convolutional Neural Network

In the recent advancements in image and video analysis, the detection of salient regions in the image becomes the initial step. This plays a crucial role in deciding the performance of such algorithms. In this work, a Multi-Resolution Feature Extraction (MRFE) technique that makes use of Discrete Wavelet Convolutional Neural Network (DWCNN) for generating features is employed. An Enhanced Feature Extraction (EFE) module extracts additional features from the high level features of the DWCNN, which are used to frame both channel as well as spatial attention models for yielding contextual attention maps. A new hybrid loss function is also proposed, which is a combination of Balanced Cross Entropy (BCE) loss and Edge based Structural Similarity (ESSIM) loss that effectively identifies and segments the salient regions with clear boundaries. The method is tested exhaustively with five different benchmark datasets and is proved superior to the existing state-of-the-art methods with a minimum Mean Absolute error (MAE) of 0.03 and F-measure of 0.956.     

基于光谱尺度空间与管道滤波的红外目标检测

为了准确地从复杂干扰背景下检测出真实弱小目标，本文引入视觉显著性，设计了基于快速光谱尺度空间与动态管道滤波的红外目标检测算法。基于真实目标与背景内容之间的整体差异，引入快速光谱尺度空间与阈值分割技术，设计视觉显著性机制，对红外图像完成处理，输出全局显著性映射，以高效过滤干扰背景内容。考虑目标与背景的局部特征差异，构建自适应局部对比度增强机制，对粗检测结果实施处理，获取对应的局部显著性映射，改善视觉显著性区域内目标的对比度。引入高斯差分理论，通过估算每一帧红外图像中的目标像素直径，形成动态管道滤波，充分消除虚警，准确识别出弱小目标。多组实验数据显示：较已有的红外目标检测技术而言，在各种不同的复杂背景干扰下，所提算法呈现出更好的检测能力，拥有更理想的接收机工作特性ROC曲线。     

基于多尺度的贝叶斯模型显著性检测

针对传统基于贝叶斯模型的显著性检测算法存在准确率不理想的问题,提出了一种基于多尺度的贝叶斯模型显著性检测算法。通过超像素分割算法(SLIC)将原图分割成不同尺度的超像素,根据超像素边界信息得到背景种子,进而通过距离计算和多尺度融合得到背景先验;对原图进行颜色增强,采用Harris算子对增强图进行检测角点求得凸包,融合不同尺度下的超像素得到凸包先验;融合背景先验和凸包先验得到最终先验;利用颜色直方图和凸包计算似然概率;将最终先验和似然概率通过贝叶斯模型计算显著图。在公开数据集MSRA1000、ECSSD上与多种传统算法进行准确率和召回率对比,该算法有更好的表现。     

结合边缘特征先验引导的深度卷积显著性检测

针对当前基于深度学习的显著性检测算法缺少利用先验特征和边缘信息，且在复杂场景中难以检测出鲁棒性强的显著性区域的问题，提出了一种结合边缘特征，利用先验信息引导的全卷积神经网络显著性检测算法。该算法利用三种被经常用到的先验知识结合边缘信息形成先验图，通过注意力机制将提取的先验特征与深度特征有效融合，最终通过提出的循环卷积反馈优化策略迭代地学习改进显著性区域，从而产生更可靠的最终显著图预测。经过实验定性定量分析，对比证明了算法的可靠性。     

观看经度联合加权全景图显著性检测算法

目的 目前针对全景图显著性检测的研究已经取得了一定成果,但在全景图像的位置特性问题中,大都仅探讨了纬度对全景图显著性检测的影响。而人们观看全景图像时,因视角有限,不同经度位置的显著性也有很大差异,从而导致预测的显著区域往往不够精确。为此本文以全景图的经度位置特性为出发点,提出基于观看经度联合加权的全景图显著性检测算法。方法 使用空间显著性预测网络得到初步的显著性图像,使用赤道偏倚进行预处理以改善不同纬度位置的显著性检测效果。接着对显著性图像进行注视点经度加权,将观察者观看全景图的行为习惯与显著性图像相结合。之后对全景图进行双立方体投影与分割,提取全景图的亮度与深度特征,进而计算不同视口经度权重。经过两次加权,得到最终的显著性图像。结果 在Salient360!挑战大赛提供的数据集上与其他几种算法进行了实验比较。结果显示,本文算法能得到很好的显著性检测结果。在对本文算法的通用性能的测试中,在标准化扫描路径显著性、相关系数、相似度与相对熵指标上分别达到了1.979 3、0.806 2、0.709 5和0.323 9,均优于其他算法。结论 提出的全景图显著性检测算法解决了以往全景图显著性检测中不同经度位置检测结果不够准确的问题。     

基于视觉显著模型的音视频交叉水印算法

提出一种基于Itti视觉显著模型的音视频交叉水印算法.首先利用音频码流的低通幅值特性生成水印,并嵌入到视频码流中;其次解码视频码流的关键帧Ⅰ帧,获取YUV序列上的亮度(Y)分量,并对其进行显著性分析;最后分块对非显著区域进行DCT变换,并通过量化索引调制嵌入水印.实验结果表明,该算法将嵌入水印后的关键帧平均PSRN值提高了1.678 5 dB,缓解了水印不可见性和鲁棒性的矛盾,实现了音视频码流之间的交叉保护.     

视觉显著性纹理—色彩特征融合的图像目标分类

针对图像目标分类,提出了一种显著性纹理特征.考虑到显著目标图像在纹理特征表征上的优势,在目标显著性图像提取的基础上进一步提取视觉显著性纹理特征.进而将该视觉显著性纹理特征同HSV色彩特征进行融合,形成图像目标融合特征,输入至后端分类器中进行分类.多类别的交叉实验证明,基于该融合特征的目标分类方法能够较为准确的对图像目标进行分类,在SIMPLIcity图像数据集上平均分类正确率达到84.84％,在Corel图像集上平均分类正确率为85.05％,优于基于单一分类特征的图像分类方法.     

一种新的基于显著图的视线跟踪方法

针对现有视线跟踪系统设备复杂、标定过程繁琐等方面的不足,提出了一种新的基于显著图的视线跟踪方法。通过红外光源设备在人眼角膜上产生的光斑中心与瞳孔中心建立瞳孔-角膜反射向量,然后将该向量作为视觉特征重构了基于显著图的视线跟踪算法。实验结果证明,提出的方法不仅缓解了视线跟踪系统标定过程繁琐的问题,而且对提高系统的精度和健壮性有一定的促进作用,这为面向人机交互的视线跟踪研究提供了可行的低成本解决方案。     

A depth perception and visual comfort guided computational model for stereoscopic 3D visual saliency

With the emerging development of three-dimensional (3D) related technologies, 3D visual saliency modeling is becoming particularly important and challenging. This paper presents a new depth perception and visual comfort guided saliency computational model for stereoscopic 3D images. The prominent advantage of the proposed model is that we incorporate the influence of depth perception and visual comfort on 3D visual saliency computation. The proposed saliency model is composed of three components: 2D image saliency, depth saliency and visual comfort based saliency. In the model, color saliency, texture saliency and spatial compactness are computed respectively and fused to derive 2D image saliency. Global disparity contrast is considered to compute depth saliency. Particularly, we train a visual comfort prediction function to distinguish stereoscopic image pair as high comfortable stereo viewing (HCSV) or low comfortable stereo viewing (LCSV), and devise different computational rules to generate a visual comfort based saliency map. The final 3D saliency map is obtained by using a linear combination and enhanced by a “saliency-center bias” model. Experimental results show that the proposed 3D saliency model outperforms the state-of-the-art models on predicting human eye fixations and visual comfort assessment.     

Image saliency estimation via random walk guided by informativeness and latent signal correlations

Visual saliency is an effective tool for perceptual image processing. In the past decades, many saliency models have been proposed by primarily considering visual cues such as local contrast and global rarity. However, such explicit cues derived only from input stimuli are often insufficient to separate targets from distractors, leading to noisy saliency maps. In fact, the latent cues, especially the latent signal correlations that link visually distinct stimuli (e.g., various parts of a salient target), may also play an important role in saliency estimation. In this paper, we propose a graph-based approach for image saliency estimation by incorporating both explicit visual cues and latent signal correlations. In our approach, the latent correlations between various image patches are first derived according to the statistical prior obtained from 10 million reference images. After that, the informativeness of image patches and their latent correlations are jointly considered to construct a directed graph, on which a random walking process is performed to generate saliency maps that pop-out only the most salient locations. Experimental results show that our approach achieves impressive performances on three public image benchmarks.