来自：http://graphics.cs.cmu.edu/projects/scene-completion/ Scene Completion Using Millions of Photographs People James Hays Alexei Efros Abstract What can you do with a million images? In this paper we present a new image completion algorithm powered by a huge database of photographs gathered from the Web. The algorithm patches up holes in images by finding similar image regions in the database that are not only seamless but also semantically valid. Our chief insight i...

原文：http://cg.cs.tsinghua.edu.cn:8080/cmm/?page_id=155 PhotoSketch: Internet Image Montage Tao Chen 1     Ming-Ming Cheng 1     Ping Tan 2     Ariel Shamir 3     Shi-Min Hu 1 1 TNList, Department of Computer Science and Technology, Tsinghua University 2 National University of Singapore     3 The Interdisciplinary Center Abstract: We present a system that composes a realistic picture from a simple freehand sketch annotated with text labels. The composed picture is gener...

http://www.keyanshequ.com/thread-471-1-1.html http://www.keyanshequ.com/redirect.php?tid=471&goto=lastpost#lastpost

What kinds of query are users likely to put to an image database? To answer this question in depth requires a detailed knowledge of user needs – why users seek images, what use they make of them, and how they judge the utility of the images they retrieve. As we show in section 3.3 below, not enough research has yet been reported to answer these questions with any certainty. Common sense evidence suggests that still images are required for a variety of reasons, including: ...

http://hi.baidu.com/wang%5Fqingy ... 46082e43a75b0a.html Picitup是一个图片搜索引擎, 很简单, 输入关键词搜索, 似乎没什么好说的. 但它的搜索过程真的是让人觉得非常专业, 比如我输入”android”进行图片搜索, 可以在搜索结果页通过”面部”, “风景”, “产品”和”颜色”来进行过滤.在过滤后的图片搜索结果页面又可以通过点击某张图片下面的”Similar Pictures”进行再次过滤。 CreativSpace也是一个基于 Google Search API图片搜索引擎，即这里搜索出来的图片和Google Images Search[Google图像搜索]是一致的。但是，Cr...

经过Canny算子边界提取后的图像，对于边界点求方向角，公式如下： 其中g(i,j)为经过高斯滤波后图像的灰度。以5度为单位，将360度的角度空间量化为72级，计算每个边界点处法向量的方向角分别落在这72级空间中的频率，这样便得到了边界方向直方图H。 边界方向不受图像中对象的位置的影响，为了达到不受图像缩放的影响，对得到的边界方向直方图进行归一化处理：H=H/S。其中H[I]为边界方向直方图，S为图像的面积。 边界方向直方图很显然受到图像旋转的影响，为了减少图像旋转对于直方图的影响，我们对直方图进行平滑：...

SIFT: Scale Invariant Feature Transform by Utkarsh in Computer Vision,Featured Matching features across different images in a common problem in computer vision. When all images are similar in nature (same scale, orientation, etc) simple corner detectors can work. But when you have images of different scales and rotations, you need to use the Scale Invariant Feature Transform. Why care about SIFT SIFT isn’t just scale invariant. You can change the following, and still...