二代提升法实现的haar小波变换程序

浙江大学的3D重建的论文，中文的，包括硕士论文和博士论文，对于研究3D重建的朋友或许有点帮助，谢谢(3D reconstruction of Zhejiang University)

K-D树源码，不错的空间查找算法，在三维重建和匹配中应用较多！(KD tree source, good space search algorithm, in the three-dimensional reconstruction and matching the application of more!)

立体匹配 asw算法实现 立体匹配经典局部算法(adapsive support weight algrithom)

实现图像纹理特征的提取，并且完成了该项功能(The extraction of image texture features, and completed the)

说明：  在图像上画一条线，提取所在位置的像素。主要是用来观察横向异质性变化的。(Draw a line on the image and extract the pixels at the location. Mainly used to observe changes in lateral heterogeneity.)

DIBR算法的简单实现，针对单幅图像，稍加改进就可以处理视频。最终生成左右分离格式的3D图像(simple implementation of DIBR，based on one frame，and can process video if the code is improved. The result is stereoscopic image.)

对图像进行欠采样，然后利用OMP 进行重构，其中用到了小波矩阵进行稀疏(The image is undersampled and then reconstructed by OMP, in which the wavelet matrix is used for sparsity.)

欧洲winner2的官方3D天线信道模型源程序 内包含十多种场景选择 考虑了角度扩展 k参数等(H=WIM(WIMPAR,LINKPAR,ANTPAR) is a 5D-array of channel coefficients. For explanation of the input parameter structs, see WIMPARSET, LAYOUT2LINK, LAYOUTPARSET, and ANTPARSET. H is a cell array of size K (number of links.SIZE(H{i})=[U S N T], where U is the number of MS (RX)elements, S is the number of BS (TX) elements, N is the number of paths,T is the number of time sample)

视频目标跟踪。在OpenCV下开发，使用meanshift方法。(video tracking. The OpenCV under development, use meanshift methods.)

实现一个心行图案的各种二维几何变换（包括同比变换，错切变换，平移变换、旋转变换等）(achieve a mind trip of two-dimensional pictorial geometry transformation (including an transform, shear transformation, translation transform, ROTATION))