基于模糊量化和2 bit深度像素的运动估计算法

提出了一种2 bit深度像素的运动估计算法。首先,将像素深度的降采样过程形式化为区间分划和区间映射2个步骤,其中前者为多对一映射,决定着运动估计性能,后者为一一映射;其次,提出一种非均匀量化方法求解区间分划的3个初始阈值,并利用隶属度函数对初始阈值细化,从而克服信号噪声等因素导致的初始阈值周围像素值的误匹配;再次,讨论了适用于2 bit深度像素运动估计的误差度量准则,进而提出了基于模糊量化和2 bit深度像素的运动估计算法;最后,借助信号自相关函数,建立比特深度转换误差—运动向量精度模型来估计该算法所能达到的预测精度。实验结果证明,对于多种类型的视频序列,尤其是场景细节和物体运动比较复杂者,该算法始终能保持较高的估计精度,运动补偿的平均峰值信噪比较之传统2 bit深度像素的运动估计提高0.27 dB。     

H．264中亚像素运动估计算法的优化

通过对H．264参考模型JM61e中运动估计算法进行分析，对整像素、1／2像素和1／4像素搜索提出了改进策略。采用自适应阈值搜索停止算法，减少了运动估计的搜索次数，且保持运动估计的精度不变。采用自适应阈值搜索停止算法，减少了运动估计搜索次数，并且基本保持运动估计的精度不变。实验结果表明，该策略在重建图像质量略有下降的情况下明显提高了编码速度。     

多传感器多尺度图像信息融合算法

在已获得对同一目标场景的多个传感器观测图像的情况下,本文建立了一种基于概率模型的多尺度图像信息融合算法.其基本思想是:首先对每个传感器图像分别进行小波包多尺度分解变换,建立基于该传感器图像的塔式结构子图像集,并且在每个尺度上得到基于每个子图像像素的概率模型;然后在每个尺度上的对应像素处,基于来自不同传感器图像的多个对应像素值,利用最小二乘规则对多尺度概率模型中的参数进行估计;再后是根据贝叶斯规则对该像素处的像素值进行融合估计;最后通过利用小波包多尺度逆变换,获得目标场景基于多个传感器图像的融合估计结果.应用该算法我们对获得的可见光和红外两种传感器图像进行计算机仿真实验,结果表明,与相关的方法相比新算法更有效.     

一种利用符号算法的自适应图象位移估计算法

本文提出了一种利用符号算法(Sign Algorithm)的自适应图象位移估计算法,这种迭代算法能直接估计出运动图象序列中两祯目标图象的位移。本文还对算法的性能进行了详细的理论分析,得出估计方差与信噪比、图象相关函数等参数之间的定量关系式,以及由于坐标量化而引起的算法对图象特征的限制条件。理论分析和模拟表明,它具有较强的抗干扰能力和计算量少,便于实时实现等优点,可应用于成象跟踪和图象编码等领域。     

小数像素运动估计快速算法

高精度的匹配和补偿可以减少预测误差，提高压缩视频图像的质量。提出了一种小数像素精度视频运动估计的快速算法，根据半像素精度运动估计的中间结果，直接推算出最高精度的运动估计。根据该算法可以快速得到任意小数像素精度的运动估计。     

AVS视频编码中子像素运动估计算法的优化研究

由于在AVS技术标准的帧间预测方案中提供了1／4子像素精度的运动估计,这部分运算量较大,运行时间长,有进一步改进的必要。本文通过对AVS视频编码标准中子像素运动估计算法的分析,针对子像素运动估计算法运算量较大的问题,提出了一种新的基于整像素SAD的快速子像素运动估计算法,它对现有的AVS技术标准子像素运动估计算法进行了较大的改进与优化。实验结果证明该方法在保持相同性能的前提下,较大程度地降低了AVS技术标准运动估计算法的复杂度。     

一种基于硬件实现的快速运动估计半像素级搜索算法

研究了在绝对差和准则下的整像素级块匹配和半像素级块匹配的联系，对绝对差和进行了合理的数学曲线拟合，通过数学曲线来预测最小绝对差和所在半像素位置，从而得到半像素级最佳匹配矢量。分析了三种不同的凹函数预测模型，提出了一种适合硬件实现的运动估计快速半像素级搜索算法。该算法直接根据整像素级运动估计的结果来推算半像素级运动估计结果，在很大程度上降低了半像素级运动估计的运算复杂度，从而利用低码率视频编码的实时实现。试验结果表明该算法可获得较好的重建图像质量。该算法利于硬件实现，可以方便地集成到现有的视频编码器中，具有较好的实用价值。     

一种基于H．264的有效运动估计算法

运动估计算法是实时视频编解码技术的研究重点，高精度的匹配和补偿可以减少预测误差，提高视频图像的压缩效果．为降低在视频编码标准H．264中运动估计的高计算复杂度问题，提出了采用一种基于节点模型的可变形块匹配运动估计算法来搜索最佳运动矢量．该算法充分利用了H．264运动矢量的的统计特性和相关性，并采用基于像素差值分类的运动估计匹配准则．实验表明，在编码性能损失很小的条件下，该算法有效降低了视频压缩编码中运动估计的运算复杂度．     

分数像素快速块匹配运动估计方法综述

介绍了视频压缩中分数像素快速块匹配运动估计的基础原理,由于全搜索算法计算量很大,需要发展快速算法。对现有快速算法进行了研究总结,介绍了所应用的数学模型、向量预测、搜索优化、提前终止4个关键技术及代表算法。最后对分数像素快速块匹配运动估计方法进行了总结和展望。     

基于H.264的内容自适应分数像素运动估计算法

分数像素精确运动估计的改进是整个运动估计模块优化的关键,本文提出了基于H.264的内容自适应分数像素运动估计算法。首先,提出基于平坦区域宏块预测的无效分数像素运动矢量(MV)搜索省略算法(SMBP);然后,改进H.264采用的基于中心的分数像素搜索算法(CBFPS),提出基于预测矢量的增强型菱形模板(EDSP)搜索算法。实验结果表明,内容自适应分数像素运动估计算法比分数像素全搜索算法(FFPS)在峰值信噪比(PSNR)有微小降低(0.095～0.209 dB)的情况下,平均减少了75.6%的分数像素搜索点,整个运动估计模块平均节省了38.5%的计算量。     

Disparity field and depth map coding for multiview 3D image generation

In the present paper techniques are examined for the coding of the depth map and disparity fields for stereo or multiview image communication applications. It is assumed that both the left and right channels of the multiview image sequence are coded using block- or object-based methods. A dynamic programming algorithm is used to estimate a disparity field between each stereo image pair. Depth is then estimated and occlusions are optionally detected, based on the estimated disparity fields. Spatial interpolation techniques are examined based on the disparity/depth information and the detection of occluded regions using either stereoscopic or trinocular camera configurations. It is seen that the presence of a third camera at the transmitter site improves the estimation of disparities, the detection of occlusions and the accuracy of the resulting spatial interpolation at the receiver. Various disparity field and depth map coding techniques are then proposed and evaluated, with emphasis given to the quality of the resulting intermediate images at the receiver site. Block-based and wireframe modeling techniques are examined for the coding of isolated depth or disparity map information. Further, 2D and 3D motion compensation techniques are evaluated for the coding of sequences of depth or disparity maps. The motion fields needed may be available as a byproduct of block-based or object-based coding of the intensity images. Experimental results are given for the evaluation of the performance of the proposed coding and spatial interpolation methods.     

Human visual system and segment-based disparity estimation

In this paper, an efficient segment-based disparity estimation algorithm is proposed, based on wavelet transform and human visual system. The core idea is to estimate disparity not only from the left and right images but also from their decomposed sub-bands up to a certain level. The stereo image pair is divided into the segments of homogeneous color. Instead of assigning a disparity value to each pixel inside a segment, a disparity plane is assigned to each segment and the stereo matching problem is formulated as an energy minimization problem in the segmented domain. The optimal disparity plane labeling is approximated by applying belief propagation, which assigns the corresponding disparity plane to each segment. The obtained disparity maps are merged into a single disparity map using the human visual system model. Experiments with stereo image pairs show the validity of the proposed method.     

Obtaining depth map from segment-based stereo matching using graph cuts

In the paper, the algorithm of segment-based stereo matching using graph cuts is developed for extracting depth information from the stereo image pairs. The first step of the algorithm employs the mean-shift algorithm to segment the reference image, which ensures our method to correctly estimate in large untextured regions and precisely localize depth boundaries, followed by the use of Adaptive Support Weighted Self-Adaptation dissimilarity algorithm (ASW-SelfAd) for the estimation of initial disparity. This is followed by application of Singular Value Decomposition (SVD) in solving the robust disparity plane fitting. In order to ensure reliable pixel sets for the segment, we filter out outliers which contain occlusion region through three main rules, namely; cross-checking, judging reliable area and disparity distance measurement. Lastly, we apply improved clustering algorithm to merge the neighboring segments. The geometrical relationship of adjacent planes such as parallelism and intersection is employed for determination of whether two planes shall be merged. A new energy function is subsequently formulated with the use of graph cuts for the refinement of the disparity map. Finally, the depth information is extracted from the final disparity map. Experimental results on the Middlebury dataset demonstrate that our approach is effective in improving the state of the art.     

Spotlight mode SAR stereo technique for height computation

This paper examines the feasibility of extracting three-dimensional (3-D) or topographic information in spotlight mode stereo synthetic aperture radar (SAR). A display of a SAR (intensity) image has two axes: range and cross-range. Elevated scatterers appear closer in range; this phenomenon is called radar image layover. How the height of each scatterer can be computed from the difference in its layover between two images is investigated. This is analogous to computing height from disparity distance (triangulation) in optical stereo. The same procedure can be applied on pixel by pixel basis for terrain elevation mapping. A general expression is derived for the accuracy of the height estimate as a function of the range resolution and the angular difference between the image planes. Accuracy increases as the angle between the image planes increases, but the bright scatterers in one image tend to fade in the other image. This limited angular persistence of radar scatterers is also discussed. Trajectories for data collection are examined that provide near-optimal height estimates while eliminating the scatterer persistency problem.     

微型实时多目立体视觉机设计与实现

本文给出一种实时计算场景稠密深度图的多目立体视觉机设计与实现方法.立体视觉机使用多个微型摄像机同步获取场景图像,采用图像修正、LoG滤波、多立体图像对匹配和稠密深度图等并行算法,利用FPGA的大规模并行处理能力和各算法间的多级流水线关系,在一片FPGA芯片上实现了立体视觉信息的实时处理.设计的立体视觉机体积小,运行速度快.当图像分辨率为320×240像素,深度搜索范围为64级,深度图精度为8位,时钟频率为60MHz时,恢复稠密深度图的速度大于30帧/秒.     

Robust ego-motion estimation and 3-D model refinement using surface parallax.

We present an iterative algorithm for robustly estimating the ego-motion and refining and updating a coarse depth map using parametric surface parallax models and brightness derivatives extracted from an image pair. Given a coarse depth map acquired by a range-finder or extracted from a digital elevation map (DEM), ego-motion is estimated by combining a global ego-motion constraint and a local brightness constancy constraint. Using the estimated camera motion and the available depth estimate, motion of the three-dimensional (3-D) points is compensated. We utilize the fact that the resulting surface parallax field is an epipolar field, and knowing its direction from the previous motion estimates, estimate its magnitude and use it to refine the depth map estimate. The parallax magnitude is estimated using a constant parallax model (CPM) which assumes a smooth parallax field and a depth based parallax model (DBPM), which models the parallax magnitude using the given depth map. We obtain confidence measures for determining the accuracy of the estimated depth values which are used to remove regions with potentially incorrect depth estimates for robustly estimating ego-motion in subsequent iterations. Experimental results using both synthetic and real data (both indoor and outdoor sequences) illustrate the effectiveness of the proposed algorithm.     

基于动态随机点立体图的立体视觉深度运动特性分析

利用动态随机点立体图,排除心理暗示线索对深度运动的影响,对立体视觉深度运动特性进行了研究.利用主观实验,分析了深度运动物体的运动位置、运动方向和运动速度等运动特性对于立体视觉舒适度(Stereo visual comfort,SVC)的影响, 并提出深度运动情况下的SVC与运动速度的预测模型.实验结果表明,动态随机点立体图中,运动物体的深度运动诱发了透视效果;在交叉视差(屏幕前)和非交叉视差(屏幕后)情况下,立体视觉深度运动导致的SVC不同.对计算得到的SVC的预测值和实测值进行相关性检测,三种深度运动的皮尔森相关系数(Pearson correlation coefficient,PCC)分别为0.956、0.972、0.977,充分表明该预测模型能够准确地预测SVC.     

Stereoscopic ranging by matching image modulations

We apply an AM-FM surface albedo model to analyze the projection of surface patterns viewed through a binocular camera system. This is used to support the use of modulation-based stereo matching where local image phase is used to compute stereo disparities. The local image phase is an advantageous feature for image matching, since the problem of computing disparities reduces to identifying local phase shifts between the stereoscopic image data. Local phase shifts, however, are problematic at high frequencies due to phase wrapping when disparities exceed +/-pi. We meld powerful multichannel Gabor image demodulation techniques for multiscale (coarse-to-fine) computation of local image phase with a disparity channel model for depth computation. The resulting framework unifies phase-based matching approaches with AM-FM surface/image models. We demonstrate the concepts in a stereo algorithm that generates a dense, accurate disparity map without the problems associated with phase wrapping.     

Inverse tensor transfer with applications to novel view synthesis and multi-baseline stereo

A new tensor transfer-based novel view synthesis (NVS) method is proposed in this paper. As opposed to conventional tensor transfer methods which transfer the pixel from the real input views to the virtual novel view, our method operates inversely in the sense that it transfers a pixel from the novel view image back to the real images. This inverse tensor-transfer approach offers a simple mechanism for associating corresponding image points across multiple views, resulting in geometrically consistent pixel chains across the input images. A colour consistency metric is used to choose the most likely colour for a pixel in the novel image by analysing the spread of colours in each of the possible pixel chains. By emphasizing colour consistency, rather than depth, our method avoids the need to precompute a dense depth map (which is essential for most conventional transfer methods), therefore alleviating many common problems with conventional methods. Experiments involving NVS on real images give promising results. The synthesized novel view image is not only photo-realistic but also has the right geometric relationship with respect to the other views.Since this method avoids explicit depth map computation, we further investigate its applicability to the multi-baseline stereo matching problem (MBS). By using this inverse transfer idea, we are able to handle non-ideally configured MBS in a natural and efficient way. The new MBS algorithm has been used for stereo vision navigation.     

基于视差估计的中间视合成

提出了一种基于视觉估计的中间视合成方法.为了生成高质量的中间视图,在视差估计中提出了一种新的基于自适应权值的方法,并在匹配的代价函数中引入视差平滑性约束项.在获得可靠密集视差图后,通过寻找中间视图中像素点在左、右图像中的对应点来合成中间视.实验结果表明,本算法可以获得很好的视差图,合成的中间视图具有较高的质量.