一种基于均值漂移的视频场景检测方法

提出了一种高效的视频场景检测方法。首先基于均值漂移,在滑动镜头窗内对各镜头聚类,并获得相应的聚类中心,然后根据电影视频场景的发展模式,计算两个镜头类之间的时序距离,接着基于时空关系进行场景检测,并且由相应的聚类中心获得场景关键帧,最后对场景过分割进行后续处理。实验证实该方法能快速聚类,并且有效地检测出场景和场景关键帧。     

基于稀疏超采样的时间性反走样算法

针对时间性反走样算法在处理帧间复用时,若场景中有较多高频颜色区域或精细模型会造成重 影、模糊、闪烁及子像素细节丢失的问题,提出了基于稀疏超采样的时间性反走样算法。基本思想是,在时间 性反走样算法的基础上,对于无法复用历史帧像素,重新引入空间域的超采样,利用剔除算法以避免不必要的 绘制开销,实现对场景的稀疏超采样。实验结果表明,该算法能够得到与超采样算法媲美的反走样效果,并具 有更高的渲染效率,有效避免重影、模糊、闪烁及子像素细节丢失的问题。      

视频标准帧速率上变换的自适应运动补偿方案

提出了一种自适应的运动补偿方案，首先利用预测三步搜索算法进行运动估计，由于此运动估计算法能够很好地利用运动向量的时空相关性．从而能得到更加平滑的运动向量场。其次．为了能够抑制运动向量场中个别奇异向量以及块匹配算法所固有的块效应问题，对向量场进行中值滤波，并采用了一种特殊的向量分配算法。方案的最后一个环节是运动补偿插值，它综合考虑插值帧的局部图像质量以及全局图像质量，利用一种白适应性较强的中值滤波操作来达到一定的综合效果。实验结果证明，此算法与运动补偿时间插值(MCTI)算法相比，插值得到的图像更加平滑，而且，两者的信噪比指标对比也充分说明了本算法的优越性。     

无线多媒体传感器网络中基于多类传感器信息的分布式视频处理

无线多媒体传感器网络中的多媒体传感器节点的功率、计算和存储能力及带宽资源受限,迫切需要低复杂度的视频处理技术对大数据量的视频进行压缩传输。基于来自网络中监控同一目标场景的多种类型传感器的信息,给出了一个新的无线多媒体传感器网络分布式视频处理系统框架;提出了一种基于多类型传感数据融合和多视角的GOP（group of picture）划分方法;在解码端,考虑单一视角视频序列之间较强的时间相关性,产生时间相关的边信息;利用来自多个视频传感节点的视频序列间的空间多视角相关性,产生多视角相关的边信息,并提供两种边信息的融合和选择机制,提高边信息的准确度和可靠性。最后仿真实验结果表明该方法的有效性和优越性。     

基于H.264的误码隐藏技术

为了在高误码环境下提高视频通信质量,提出一种改进的基于H.264的误码隐藏技术。对于帧内编码图像,根据图像直方图信息计算图像的光滑特性,然后自适应地选择线性插值的方法。对于帧间编码图像,根据不同大小块编码类型和多参考帧的特性,能够更准确地估计出丢失的MV,进行时域误码隐藏。实验表明,此方法同H.264参考软件JM11.0方法相比,具有更好的主客观图像恢复质量。     

基于光流场动态参与介质绘制算法

为了解决动态参与介质场景连续帧的实时绘制问题,提出了一种基于光流场动态参与介质场景的绘制算法。首先,采用区域匹配的方法计算关键帧之间的光流场;然后,通过插值的方法计算中间帧之间的光流场,采用帧间光流连贯性函数表示帧与帧之间光流一致性的度量,保证帧与帧之间介质运动不会发生突变;最后,按照所产生的光流场,绘制连续动态变化的参与介质场景。在连续5帧动态参与介质场景绘制中,所提算法比基于径向基函数模型(RBF)参与介质的光子映射算法效率提高了近3倍,能够达到连续帧的实时绘制,且绘制质量比较高。     

一种适用于H.264的时域差错掩盖算法

为了克服视频传输中因传输错误引起的视频质量下降,结合H.264的编码特性,提出一种基于运动矢量自适应修复的时域差错掩盖算法。首先利用正确接收的宏块信息估计丢失块邻域内运动矢量的变化强度,据此自适应采用边界运动矢量差值最小法或多项式插值法恢复丢失块的运动矢量,然后根据恢复的运动矢量对丢失块进行补偿替代。仿真结果表明,该算法同传统的时域差错掩盖算法相比,在相同网络丢包的环境下PSNR值提高了0.2~2.5dB,同时解码的图像也获得较好的主观视觉效果。     

Novel error concealment method with adaptive prediction to the abrupt and gradual scene changes

In this paper, the impact of scene change on the conventional error concealment method is addressed and a novel error concealment method is proposed to improve the insufficiency of the conventional temporal error concealment algorithm due to the occurrence of scene change. Combining with the low complexity scene change detection algorithm using macroblock type information, the corrupt blocks resulting from bit errors are concealed either temporally or spatially depending on whether or not an abrupt scene change is found. In the case of gradual scene change, a novel error concealment method of interpolation and extrapolation is proposed to utilize the linear property of gradual scene change sequence, and effectively reduce the concealment error in comparison with the conventional algorithm. A great improvement of about 3 to 5 dB PSNR on average and 6 to 8 dB in some cases is obtained with very little memory and computation overhead.     

高效的I帧分区错误隐藏方法

针对现有I帧错误隐藏方法不能平衡恢复图像质量与算法复杂度的问题,提出了一种高效的I帧分区错误隐藏方法。首先,利用视频帧之间的运动相关性将丢失宏块分为运动宏块和静止宏块。对于静止宏块,采用帧拷贝法进行掩盖;对于运动宏块,再根据其周围正确解码宏块的纹理信息将其分为平滑块和纹理块。对平滑块采用双线性插值法进行恢复;对纹理块利用比较精细的指数分布权重的加权模板匹配(WTE)法进行掩盖。实验结果表明,与WTE算法相比,所提方法的峰值信噪比(PSNR)平均提高了2.6 dB,计算复杂度平均降低了90%。对于场景连续的具有不同特征和分辨率的视频序列,所提方法都具有一定的适用性。     

A flexible side information generation framework for distributed video coding

One of the most efficient approaches to generate the side information (SI) in distributed video codecs is through motion compensated frame interpolation where the current frame is estimated based on past and future reference frames. However, this approach leads to significant spatial and temporal variations in the correlation noise between the source at the encoder and the SI at the decoder. In such scenario, it would be useful to design an architecture where the SI can be more robustly generated at the block level, avoiding the creation of SI frame regions with lower correlation, largely responsible for some coding efficiency losses. In this paper, a flexible framework to generate SI at the block level in two modes is presented: while the first mode corresponds to a motion compensated interpolation (MCI) technique, the second mode corresponds to a motion compensated quality enhancement (MCQE) technique where a low quality Intra block sent by the encoder is used to generate the SI by doing motion estimation with the help of the reference frames. The novel MCQE mode can be overall advantageous from the rate-distortion point of view, even if some rate has to be invested in the low quality Intra coding blocks, for blocks where the MCI produces SI with lower correlation. The overall solution is evaluated in terms of RD performance with improvements up to 2 dB, especially for high motion video sequences and long Group of Pictures (GOP) sizes.     

一种时空联合的视频运动目标提取与跟踪新算法

提出了一种时空联合的视频运动目标提取与跟踪新算法。在空域分割中,针对分水岭方法过分割现象明显的缺点,对分水岭分割方法进行了改进;在时域分割中,首先对全局运动进行了补偿,随后为消除仅用两帧帧差进行对象分割所带来的误差,采用多帧帧差求和的方法,并自适应选取累积帧差的二值化阈值;时空分割结果进行投影融合后得到视频对象,提出用一种基于区域子块匹配的方法跟踪视频对象。实验结果表明,该算法简洁有效,能较好地把对象从运动背景中提取出来,并实现跟踪。     

Video scene analysis in 3D wavelet transform domain

This paper proposes a novel scene analysis algorithm based on three-dimensional discrete wavelet transform (3D DWT). Based on the correlation among the adjacent frames, video frames can be considered as four categories: abrupt scene transition, motion scene, gradual scene transition and static scene, which are ranked from low to high according to the strength of the correlation. Through the investigation of the particular temporal and spatial distribution of each category, the correlation among adjacent frames could be described by the 3D DWT coefficients related statistical features, which are the energy of high-frequency coefficients difference, the sum of high-frequency coefficients magnitudes and the difference of low-frequency coefficients magnitudes. The energy of high-frequency coefficients difference is first used to detect the abrupt scene transition including cut and flashlight. Then all the three features are input to SVM for the purpose of analyzing the residual scenes and detecting the gradual scene transition, such as dissolve and fade. Experimental results show the method to be effective not only for the abrupt scene transition detection, but also for the gradual scene transition detection.     

Fast scene change detection using direct feature extraction fromMPEG compressed videos

In order to process video data efficiently, a video segmentation technique through scene change detection must be required. This is a fundamental operation used in many digital video applications such as digital libraries, video on demand (VOD), etc. Many of these advanced video applications require manipulations of compressed video signals. So, the scene change detection process is achieved by analyzing the video directly in the compressed domain, thereby avoiding the overhead of decompressing video into individual frames in the pixel domain. In this paper, we propose a fast scene change detection algorithm using direct feature extraction from MPEG compressed videos, and evaluate this technique using sample video data, First, we derive binary edge maps from the AC coefficients in blocks which were discrete cosine transformed. Second, we measure edge orientation, strength and offset using correlation between the AC coefficients in the derived binary edge maps. Finally, we match two consecutive frames using these two features (edge orientation and strength). This process was made possible by a new mathematical formulation for deriving the edge information directly from the discrete cosine transform (DCT) coefficients. We have shown that the proposed algorithm is faster or more accurate than the previously known scene change detection algorithms     

基于时空相关性的分布式压缩感知多假设预测重构算法

综合考虑到视频序列本身的不同特性以及时空相关性, 将传统视频编码中的多假设预测运动估计思想引入到分布式压缩感知视频编码系统中, 提出一种新的基于时空相关性的分布式压缩感知多假设预测重构算法。在编码端增加CS帧的块模式判别, 在解码端CS帧根据模式判别进行相应的基于块的多假设预测估计和残差稀疏重构。仿真实验表明, 与传统的分布式压缩感知多假设预测重构算法相比, 所提出的算法能够较好地改善视频序列的重建质量。     

基于参考帧的视频播放超分辨率放大算法

利用连续视频各帧之间的时域和空域相关性,在视频播放超分辨率插值放大重建时,主要对I帧进行边缘检测插值,而对P帧和B帧图像的插值则参考I帧的插值结果,从而提高整个视频播放的主观质量,计算量也不大.实验结果表明,本算法对提高每帧图像以及整体视频主观质量均有明显效果,大大减少了播放时由于视频的放大而出现的块状和锯齿状等失真.     

An adaptive image interpolation algorithm for image/video processing

Image interpolation is one of the key technologies in image/video processing. In this study, a new adaptive image interpolation algorithm is proposed. The objective of the proposed approach is to recover up-sampled image frames from the corresponding decimated (low-resolution) image frames. In the proposed approach, within each iteration, two proposed nonlinear filters are utilized to iteratively generate high-frequency components lost within the decimation procedure. Finally, a post-processing procedure is adopted to reduce the blocking artifacts within the interpolated images. Based on the experimental results obtained in this study, in terms of the average PSNR_p (peak signal-to-noise ratio) in dB and subjective measure of the quality of the interpolated images, the interpolation results by the proposed approach are better than that by three existing interpolation approaches for comparison.     

低小慢无人机降落野外场景识别方法

针对低小慢无人机野外飞行场景复杂自主降落场景识别问题,提出了一种融合局部金字塔特征和卷积神经网络学习特征的野外场景识别算法。首先,将场景分为4×4和8×8块的小场景,使用方向梯度直方图（HOG）算法提取所有块的场景特征,所有特征首尾连接得到具有空间金字塔特性的特征向量。其次,设计一个针对场景分类的深度卷积神经网络,采用调优训练方法得到卷积神经网络模型,并提取深度网络学习特征。最后,连接两个特征得到最终场景特征,并使用支持向量机（SVM）分类器进行分类。所提算法在Sports-8、Scene-15、Indoor-67以及自建数据集上较传统手工特征方法的识别准确率提高了4个百分点以上。实验结果表明,所提算法能有效提升降落场景识别准确率。     

Spatial?temporal content-adaptive deinterlacing algorithm

An algorithm, which adapts to an appropriate interpolation technique for high-quality deinterlacing based on the spatial?temporal spectral and edge-oriented features of local video contents, is introduced. This algorithm employs a spectrum-adaptive vertical?temporal filter to deal with generic video scenes, and adopts a texture-adaptive edge-based line-averaging interpolation with an edge-consistency check to interpolate moving edges with high efficiency. This deinterlacing algorithm likewise incorporates an economical recursive stationary-pixel detection scheme with a field-averaging filter to effectively interpolate stationary video scenes. Because of the precise selection of the interpolation technique for versatile video contents, the proposed algorithm provides not only better objective peak signal-to-noise ratio, but also results in more impressive subjective visual quality when compared with other non-motion compensated deinterlacing approaches.     

相似性约束人脸区域的H.264视频流自适应错误隐藏研究

针对H.264视频压缩码流在信道传输过程中易发生丢包和误码错误,以及视频图像解码重建后人脸区域模糊等问题,提出了一种基于相似性约束人脸区域的自适应错误隐藏算法,对错误块进行两次错误隐藏。首先利用自适应方向插值空域错误隐藏算法对错误块中每个像素点根据其纹理方向进行自适应方向插值运算并恢复,实现第一次错误隐藏,然后提取错误块视频图像的前景目标,采用肤色分割的方法定位出可能的人脸区域,再基于相似性约束方法确定用于覆盖错误块人脸区域的目标人脸区域,通过仿射变换后,将目标人脸区域的梯度信息覆盖错误块的人脸区域,实现第二次错误隐藏。改进的方法能够更加精确的确定插值方向且更好的恢复人脸区域,重建后的视频平滑性有一定的改善。采用JM8.6平台仿真结果表明,与现有基于人脸五官特征的空域差错掩盖算法相比,视频图像隐藏效果在主观视觉上有一定改善,计算出的峰值信噪比(PSNR)提高了1.21dB。     

基于相关区域的H.264/AVC时域错误隐藏

在视频内容分裂处,相邻块的运动矢量会出现突变,简单利用插值恢复丢失块的运动矢量一般有较大误差。为了减少这种误差,依据丢失宏块附近块运动矢量的变化趋势,构造视频内容分裂预测曲线,划分相关区域,对每个相关区域,运用准值插值估计丢失块的运动矢量。模拟测试表明,该方法能够有效提高视频质量。