【摘 要】
：
Computational photography is an emerging multidisciplinary field.Over the last two decades,it has integrated studies across computer vision,computer graphics,signal processing,applied optics and relat
【出 处】
：
Science China(Information Sciences)
论文部分内容阅读
Computational photography is an emerging multidisciplinary field.Over the last two decades,it has integrated studies across computer vision,computer graphics,signal processing,applied optics and related disciplines.Researchers are exploring new ways to
其他文献
In this paper we propose an optimized seam-carving approach for anti-shearing image resizing.Current image/video seam-carving strategy only focuses on deleting the pixels along 8-connected seams,which
Although many efficient image cutout approaches have been proposed,it is still a challenge to cut out foreground objects from a video clip because of the costs of time and memory consumption and the d
This paper presents a graphics processing unit (GPU) based stylization approach that preserves the fine structure between the original and the stylized images using gradient optimization.Existing abst
We articulate a novel approach to geometric model completion via interactive sketches in this paper.First,the initial incomplete model (with holes) is decomposed into a base model and a high-frequency
A new model-based human body tracking framework with learning-based theory is introduced in this paper.We propose a variable structure multiple model (VSMM) framework to address challenging problems s
Traditional secret sharing (SS) schemes can reconstruct the secret precisely,but have high com-putation complexity.Visual secret sharing (VSS) schemes use human visual system to reconstruct the secret
针对复杂场景中运动目标较难定位的问题,提出一种结合纹理和颜色特征的AdaBoost目标跟踪算法.首先在线训练一个弱分类器的集合区分目标和背景;然后,通过AdaBoost将集合中的各弱分类器组合成一个强分类器,用于标定下一帧中各像素的类别属性,并生成置信图;最后,在置信图中用Mean Shift算法定位目标的中心.实验结果表明,该算法在光照变化、目标自身发生形变和遮挡的情况下,能准确地对目标进行跟踪
前景/背景分割算法是计算机视觉中一种常见的算法。其基本思想是利用背景中不同像素或帧与帧之间的相关性,判断每个像素点的灰度值,然后根据预测值和实际观察值判断当前像素属于前景还是背景。首先介绍了几种应用在不同场合的前景/背景分割算法。考虑到应用传统的基于处理器的平台很难实时实现这类计算量很大的算法,所以在该算法的有效性被确认后,重点介绍其嵌入式实时实现。同时引入了一种先进的实现算法的方法:可重构计算及
提出一种抗噪声多聚焦图像融合方法.首先,采用改进拉普拉斯金字塔(LP)变换构造图像的塔形数据结构.每层数据经脉冲耦合神经网络(PCNN)迭代运算生成相应的点火次数矩阵;然后,以此为依据经判决算子完成数据融合;最后,采用伪逆重构算法生成融合图像.实验结果表明,所提出的方法具有抗加性噪声及JPEG压缩系统噪声的能力,融合图像的客观评价指标较高,且能有效减弱原LP算法融合图像边缘的"Gibss"伪影现象
针对运动声阵列在有色噪声环境中的非线性滤波跟踪问题,提出一种确定性核粒子群的粒子滤波算法.该算法通过确定性初始化核粒子集、确定性后验概率密度函数及粒子群与核粒子集更新方式来提高跟踪的精度,并推导出该算法的理论误差性能下界.与传统的粒子滤波算法相比,仿真结果表明了所提出算法的有效性和优越性.