搜索结果: 121-135 共查到“知识库 数字媒体技术”相关记录1427条 . 查询时间(4.094 秒)
基于快速SIFT特征提取的模板匹配算法
尺度不变特征变换算法 金字塔描述符 种子向量 旋转不变性 半数搜索法
2012/3/1
针对尺度不变特征变换(SIFT)算法计算量大的问题,提出一种基于快速SIFT特征提取的模板匹配算法。采用递推方法加速计算每个种子向量,利用向量排序来保持算法对旋转的不变性,并通过一种适用于该描述符的快速搜索匹配方法,提高算法的实时性。实验结果表明,该算法能提高匹配速度,并且能在旋转、缩放的情况下有效地识别目标。
提出一种三维地层剖面的实时显示方法。该方法利用给定的数字高程模型数据及三次样条函数拟合曲线构造剖面轮廓线,以有效解决三维地层模型内部的可视化问题。为便于施工,在疏浚过程中采用声纳数据动态调整模型。实验结果表明,该方法计算速度快、精度高,可以达到实时显示的效果,并能提高耙吸挖泥船疏浚的质量,减少对环境的影响。
Robust Object Tracking via Inertial Potential based Mean Shift
Object tracking mean shift motion modeling probabilistic approximation
2013/7/24
We present a novel mean shift approach in this paper forrobust object tracking based on an inertial potential model.Conventional mean shift based trackers exploit only appear-ance information of obser...
Random Partial Paired Comparison for Subjective Video Quality Assessment via HodgeRank ∗
Subjective Video Quality Assessment HodgeRank, Random Graphs Persistence Homology
2013/7/24
Subjective visual quality evaluation provides the groundtruth and source of inspiration in building objective visualquality metrics. Paired comparison is expected to yieldmore reliable results; howeve...
Visual Conspicuity Index: Spatial Dissimilarity, Distance, and Central Bias
central bias conspicuity dissimilarity spatial distance visual saliency
2013/7/24
We propose an image conspicuity index that combines three factors: spatial dissimilarity, spatial distance andentral bias. The dissimilarity between image patches is evaluatedin a reduced dimensional ...
Abstract—The linear regression model is a very attractive toolto design effective image interpolation schemes. Some regression-based imageinterpolation algorithms have been proposedin the literature, ...
Treat samples differently: Object tracking with semi-supervised online CovBoost
Object tracking semi-supervised CovBoost
2013/7/24
Most feature selection methods for object tracking assume that the labeled samples obtained in the next framesfollow the similar distribution with the samples in the previous frame. However, this assu...
Face Recognition Based on Non-Corresponding Region Matching
Recognition Non-Corresponding Region Matching
2013/7/24
In previous works of face recognition, similarity betweenfaces is measured by comparing corresponding face regions. That is to say, matching eyes with eyes and mouths withmouths etc.. In this paper, w...
Human Group Activity Analysis with Fusion of Motion and Appearance Information
Human Group Activity Activity Analysis Feature Fusion
2013/7/24
Human activity analysis is an important and challenging task in video content analysis and understanding. In this paper, we focus on the activity of small human group, which involves countable persons...
Adaptive Image Deblurring via Tanner Graph Representation and Belief Propagation
Adaptive Representation Belief Propagation
2013/7/24
Image deblurring aims to counteract the effect of point spread function (PSF) inimperfect image formation system. Conventional solutions include frequency-domainapproaches (e.g. Wiener filtering...
Salient region detection and segmentation for general object recognition and image understanding
object recognition image understanding visual saliency salient object segmentation visual dic- tionary
2013/7/24
General object recognition and image understanding is recognized as a dramatic goal for computervision and multimedia retrieval. In spite of the great e®orts devoted in the last two decades, it s...
层次化互连结构的EVMPSoC设计与实现
嵌入式可视媒体处理 层次化高低速系统总线 多通道访存 多核SoC
2012/2/13
为满足计算密集且数据带宽大的混合多媒体应用在嵌入式系统中的实现需求,介绍了一款采用层次化互连结构的异构多核嵌入式可视媒体处理系统芯片(EVMPSoC)的设计与实现方法。该SoC芯片由一个32位嵌入式RSIC主处理器EPStar 3和两个应用定制指令集的SIMD协处理器核组成,采用层次化高低速总线和多通道双位宽并行访存结构进行互连,以满足混合多媒体处理应用的通信需求。该芯片在SMIC 0.13 μm...
Generating Descriptive Visual Words and Visual Phrases for Large-Scale Image Applications
Image retrieval image search re-ranking object recognition visual phrase visual word.
2013/7/24
Bag-of-visual Words (BoWs) representation hasbeen applied for various problems in the fields of multimediaand computer vision. The basic idea is to represent images asvisual documents composed o...
Maximal Linear Embedding for Dimensionality Reduction
Dimensionality reduction manifold learning maximal linear patch landmarks-based global alignment.
2013/7/24
Over the past few decades, dimensionality reduction has been widely exploited in computer vision and pattern analysis.This paper proposes a simple but effective nonlinear dimensionality reduction algo...
SIDE INFORMATION EXTRAPOLATION WITH TEMPORAL AND SPATIAL CONSISTENCY
Distributed video coding side information extrapolation temporal and spatial consistency regularized local linear regression
2013/7/24
In this paper, we present an efficient side information extrapolation scheme with temporal and spatial consistency for lowdelay Wyner-Ziv videocoding. Our method is based on theregularized local...