图像处理和计算机视觉中的经典论文（部分）

悠悠 2022-06-13 02:37 185阅读 0赞

自己视野狭小，不敢说全部，只是把自己熟悉的方向中的部分经典文章列出来了。经典的论文，读得怎么透都不过分。有人说关于配准的文章太多了，其实我也不太关注这方面，不过由于它们引用率都比较高，就都列出来了，不过在zip包里一篇都没有。不关注这方面的可以无视之。

文档和论文下载地址：

[http://iask.sina.com.cn/u/2252291285/ish?folderid=775855][http_iask.sina.com.cn_u_2252291285_ish_folderid_775855]

# 1990年之前 #

<table> 
 <tbody> 
 <tr> 
 <td> Peter Burt, Edward Adelson. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=21318" rel="nofollow">The Laplacian Pyramid as A Compact Image Code</a>. </td> 
 <td> 虽说这个Laplacian Pyramid是有冗余的，但使用起来非常简单方便，对理解小波变换也非常有帮助。这位Adelson是W.T.Freeman的老板，都是大牛. </td> 
 </tr> 
 <tr> 
 <td> J Canny </td> 
 <td> A Computational Approach to Edge Detection </td> 
 <td> 经典不需要解释。在Sonka的书里面对这个算法也有比较详细的描述。 </td> 
 </tr> 
 <tr> 
 <td> S Mallat. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=13929" rel="nofollow">A theory for multiresolution signal decomposition: The wavelet representation</a> </td> 
 <td> Mallat的代表作。 </td> 
 </tr> 
 <tr> 
 <td> M Kass, A Witkin, D Terzopoulos. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=14062" rel="nofollow">Snakes: active contour models</a> </td> 
 <td> Deformable model的开山鼻祖。 </td> 
 </tr> 
 <tr> 
 <td> RM HARALICK </td> 
 <td> <a href="http://dceanalysis.bigr.nl/Haralick73-Textural%20features%20for%20image%20classification.pdf" rel="nofollow">Textural Features for Image Classiﬁcation</a> </td> 
 <td> 这三篇都是关于纹理特征的，虽然过去这么多年了，现在在检索和识别中依然很有用。 </td> 
 </tr> 
 <tr> 
 <td> RM HARALICK </td> 
 <td> Statistical and structural approaches to texture </td> 
 <td> &nbsp; </td> 
 </tr> 
 <tr> 
 <td> Tamura等 </td> 
 <td> Texture features corresponding to visual perception </td> 
 <td> &nbsp; </td> 
 </tr> 
 <tr> 
 <td> A P Dempster, N M Laird, D B Rubin. 1977 </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=419" rel="nofollow">Maximum likelihood from incomplete data via the EM algorithm&nbsp;</a> </td> 
 <td> EM算法在计算机视觉中有着非常重要的作用 </td> 
 </tr> 
 <tr> 
 <td> L Rabiner. 1989 </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=1069" rel="nofollow">A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition</a> </td> 
 <td> HMM同样是计算机视觉必须掌握的一项工具 </td> 
 </tr> 
 <tr> 
 <td> B D Lucas, T Kanade </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=16204" rel="nofollow">An iterative image registration technique with an application to stereo- vision</a> </td> 
 <td> Lucas </td> 
 </tr> 
 <tr> 
 <td> J R Quinlan </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary;jsessionid=6A291407B748557C73B8B018747CDE80?cid=7412" rel="nofollow">Induction of decision trees</a> </td> 
 <td> 偏模式识别和机器学习一点 </td> 
 </tr> 
 </tbody> 
</table>

# 1990年 #

<table> 
 <tbody> 
 <tr> 
 <td> P Perona, J Malik. PAMI </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=24166" rel="nofollow">Scale-space and edge detection using anisotropic diffusion</a> </td> 
 <td> 关于scale space最早的一篇论文之一，引用率很高 </td> 
 </tr> 
 <tr> 
 <td> T Lindeberg </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=102213" rel="nofollow">Scale-space for discrete signals</a>. </td> 
 <td> Lindeberg关于scale space比较早的一篇，后续还有好几篇 </td> 
 </tr> 
 <tr> 
 <td> anzad, A.; Hong, Y.H. </td> 
 <td> Invariant image recognition by Zernike moments </td> 
 <td> Zernike moment,做过模式识别或者检索的应该都知道这个东东 </td> 
 </tr> 
 </tbody> 
</table>

# 1991年 #

<table> 
 <tbody> 
 <tr> 
 <td> W Freeman, E Adelson. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=46534" rel="nofollow">The design and use of steerable filters</a> </td> 
 <td> Freeman最早的一篇力作，也是我读的第一篇学术论文。现在Freeman在MIT风生水起，早已是IEEE Fellow了 </td> 
 </tr> 
 <tr> 
 <td> Michael J. Swain , Dana H. Ballard </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=37603" rel="nofollow">Color Indexing</a>. </td> 
 <td> google scholar上引用将近五千次 </td> 
 </tr> 
 <tr> 
 <td> MA Turk CVPR </td> 
 <td> Face recognition using eigenfaces </td> 
 <td> &nbsp; </td> 
 </tr> 
 </tbody> 
</table>

# 1992年 #

<table> 
 <tbody> 
 <tr> 
 <td> L G Brown. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=34639" rel="nofollow">A survey of image registration techniques</a>. </td> 
 <td> 比较早的一篇关于配准的综述了 </td> 
 </tr> 
 </tbody> 
</table>

# 1993年 #

<table> 
 <tbody> 
 <tr> 
 <td> S G Mallat, Z Zhang. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=35168" rel="nofollow">Matching pursuits with time-frequency dictionaries</a> </td> 
 <td> Mallat另一篇关于小波的文章，不研究小波的可以无视之 </td> 
 </tr> 
 <tr> 
 <td> L Vincent. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=12926" rel="nofollow">Morphological grayscale reconstruction in image analysis: Applications and efficient algorithms</a> </td> 
 <td> &nbsp; </td> 
 </tr> 
 <tr> 
 <td> DP Huttenlocher </td> 
 <td> Comparing images using the Hausdorff distance </td> 
 <td> Google scolar上引用2200多次 </td> 
 </tr> 
 </tbody> 
</table>

# 1994年 #

<table> 
 <tbody> 
 <tr> 
 <td> J Shi, C Tomasi. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=39147" rel="nofollow">Good feature to track</a>. </td> 
 <td> Tomasi这个名字还会出现好几次，真的很牛 </td> 
 </tr> 
 <tr> 
 <td> Linderberg </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=72162" rel="nofollow">Scale-space theory in computer vision</a> </td> 
 <td> &nbsp; </td> 
 </tr> 
 <tr> 
 <td> J L Barron, D J Fleet, S S Beauchemin. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=32063" rel="nofollow">Performance of optical flow techniques</a>. </td> 
 <td> &nbsp; </td> 
 </tr> 
 </tbody> 
</table>

# 1995年 #

<table> 
 <tbody> 
 <tr> 
 <td> R Malladi, J Sethian, B Vemuri. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=18338" rel="nofollow">Shape Modeling with Front Propagation: A Level Set Approach</a> </td> 
 <td> Level set的经典文章 </td> 
 </tr> 
 <tr> 
 <td> TF COOTES </td> 
 <td> Active Shape Models-Their Training and Application </td> 
 <td> ASM </td> 
 </tr> 
 <tr> 
 <td> MA Stricker </td> 
 <td> <a href="http://scholar.google.com.hk/scholar_url?hl=zh-CN&amp;q=http://citeseerx.ist.psu.edu/viewdoc/download%3Fdoi%3D10.1.1.41.2789%26rep%3Drep1%26type%3Dpdf&amp;sa=X&amp;scisig=AAGBfm1jkSQZxKSUc36dAb55sbrkGGq2fg&amp;oi=scholarr&amp;ei=PaZYT4z5B_CziQeA8-C7DQ&amp;ved=0CCkQgAMoADAA" rel="nofollow">Similarity of color images</a> </td> 
 <td> 颜色检索相关 </td> 
 </tr> 
 <tr> 
 <td> C Cortes, V Vapnik. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=45811" rel="nofollow">Support-vector networks</a>. </td> 
 <td> SVM在计算机视觉中也有着非常重要的地位 </td> 
 </tr> 
 </tbody> 
</table>

# 1996年 #

<table> 
 <tbody> 
 <tr> 
 <td> T MCINERNEY. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=79374" rel="nofollow">Deformable models in medical image analysis: A survey</a> </td> 
 <td> 活动模型的一篇较早的综述 </td> 
 </tr> 
 <tr> 
 <td> Tai Sing Lee </td> 
 <td> Image Representation Using 2D Gabor Wavelets </td> 
 <td> Google引用也有近千次 </td> 
 </tr> 
 <tr> 
 <td> Amir Said,&nbsp; A. Pearlman </td> 
 <td> A New, Fast, and Efﬁcient Image Codec Based on Set Partitioning in Hierarchical Tree </td> 
 <td> SPIHT。图像压缩领域与EBCOT齐名的经典算法。 </td> 
 </tr> 
 <tr> 
 <td> L P Kaelbling, M L Littman, A W Moore. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=1455" rel="nofollow">Reinforcement learning: A survey</a> </td> 
 <td> 机器学习里面的一篇综述，引用率比较高，就列在这了。 </td> 
 </tr> 
 <tr> 
 <td> <a rel="nofollow">B. S. Manjunath</a>&nbsp;and W. Y. Ma </td> 
 <td> Texture features for browsing and retrieval of image data </td> 
 <td> 检索的文章比较多，其实它们的应用不仅仅是检索。只要是需要提取特征的地方，检索里面的方法都可以用到 </td> 
 </tr> 
 <tr> 
 <td> &nbsp; </td> 
 <td> comparing images using color coherence vectors </td> 
 <td> 检索中的CCV方法 </td> 
 </tr> 
 <tr> 
 <td> &nbsp; </td> 
 <td> Image retrieval using color and shape </td> 
 <td> 关于形状特征后面有一篇综述 </td> 
 </tr> 
 </tbody> 
</table>

# 1997年 #

<table> 
 <tbody> 
 <tr> 
 <td> V Caselles, R Kimmel, G Sapiro. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=40833" rel="nofollow">Geodesic active contours</a> </td> 
 <td> 活动轮廓模型的一个小分支 </td> 
 </tr> 
 <tr> 
 <td> R E Schapire, Y Freund, P Bartlett, W S Lee. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=27821" rel="nofollow">Boosting the Margin: A New Explanation for the Effectiveness of Voting Methods</a>. </td> 
 <td> Schapire和Freund发明了Adaboost，给计算机视觉带来了不少经典算法 </td> 
 </tr> 
 <tr> 
 <td> F Maes, D Vandermeulen, G Marchal, P Suetens. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=79497" rel="nofollow">Multimodality image registration by maximization of mutual information</a> </td> 
 <td> 互信息量配准 </td> 
 </tr> 
 <tr> 
 <td> E Osuna, R Freund, F Girosi. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=71777" rel="nofollow">Training support vector machines: An application to face detection</a>. </td> 
 <td> SVM在人脸检测中的应用。不过人脸检测最经典的方法应该是Viola-Jones </td> 
 </tr> 
 <tr> 
 <td> J Huang, S Kumar, M Mitra, W-J Zhu, R Zabih. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=60103" rel="nofollow">Image indexing using color correlogram</a> </td> 
 <td> Color correlogram，检索中的又一个颜色特征。和前面的CCV以及颜色矩特征基本上覆盖了所有的颜色特征。 </td> 
 </tr> 
 <tr> 
 <td> Y Freund, R Schapire. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=27811" rel="nofollow">A decisiontheoretic generalization of on-line learning and an application to boosting</a>.&nbsp; </td> 
 <td> Adaboost的经典文章 </td> 
 </tr> 
 </tbody> 
</table>

# 1998年 #

1998年是图像处理和计算机视觉经典文章井喷的一年。大概从这一年开始，开始有了新的趋势。由于竞争的加剧，一些好的算法都先发在会议上了，先占个坑，等过一两年之后再扩展到会议上。

<table> 
 <tbody> 
 <tr> 
 <td> T Lindeberg </td> 
 <td> Feature detection with automatic scale selection </td> 
 <td> Linderberg的scale space到此为止基本结束了。在一些边缘提取，道路或者血管检测中，scale space确实是一种很不错的工具 </td> 
 </tr> 
 <tr> 
 <td> C J C Burges. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=6356" rel="nofollow">A tutorial on support vector machines for pattern recognition</a>. </td> 
 <td> 使用svm的话，这篇文章应该是必读的了。比95年那篇原始文章引用率还高 </td> 
 </tr> 
 <tr> 
 <td> M Isard, A Blake. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=394" rel="nofollow">CONDENSATION – Conditional Density Propagation for Visual Tracking</a> </td> 
 <td> Tracking中的经典文章了 </td> 
 </tr> 
 <tr> 
 <td> L Page, S Brin, R Motwani, T Winograd </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=6542" rel="nofollow">The PageRank citation ranking: bringing order to the web</a> </td> 
 <td> 这篇文章应该不属于CV的范畴，鉴于作者的大名鼎鼎，暂且列在这 </td> 
 </tr> 
 <tr> 
 <td> C Tomasi, R Manduchi. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=19313" rel="nofollow">Bilateral filtering for gray and color images</a>. </td> 
 <td> 做过图像滤波平滑去噪或者HDR的应该都知道Bilateral filter。原理非常非常简单，简单到一个公式就可以概括这篇文章，简单到实在无法扩充到期刊。这也是Tomasi第二次出现了。一直很纳闷，这个很直观的思想在这之前怎么就从来没人提呢。 </td> 
 </tr> 
 <tr> 
 <td> C Xu, J L Prince. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=113219" rel="nofollow">Snakes, shapes and gradient vector flow</a>. </td> 
 <td> 终于碰到中国人写的文章了，很荣幸还是校友。GVF是snake和level set领域的重要分支和方法 </td> 
 </tr> 
 <tr> 
 <td> Wim Sweldens. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=44857" rel="nofollow">The lifting scheme: A construction of second generation wavelets</a>. </td> 
 <td> 第二代小波。真正让小波有了实用价值，在JPEG2000中就采用的提升小波。个人更喜欢的是下一篇，简单易懂，字体也大 </td> 
 </tr> 
 <tr> 
 <td> Daubechies Wim Sweldens </td> 
 <td> Factoring wavelet transforms into lifting steps </td> 
 <td> 另一位作者也很牛，小波十讲的作者 </td> 
 </tr> 
 <tr> 
 <td> H A Rowley, S Baluja, T Kanade. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=9154" rel="nofollow">Neural Network-based Face Detection</a>. </td> 
 <td> 做人脸的应该是必看的了。不做人脸的话应该可以不用看吧 </td> 
 </tr> 
 <tr> 
 <td> J B A Maintz, M A Viergever. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=117142" rel="nofollow">A survey of medical image registration</a>. </td> 
 <td> 关于图像配准的另一篇综述 </td> 
 </tr> 
 <tr> 
 <td> T F Cootes, G J Edwards, C J Taylor. </td> 
 <td> Active Appearance Models </td> 
 <td> AAM </td> 
 </tr> 
 </tbody> 
</table>

# 1999年 #

<table> 
 <tbody> 
 <tr> 
 <td> D Lowe. &nbsp; </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=37222" rel="nofollow">Object Recognition from Local Scale-invariant Features</a> </td> 
 <td> 大名鼎鼎的SIFT，后面有一篇IJCV上的Journal版本，更全面一点。 </td> 
 </tr> 
 <tr> 
 <td> R E Schapire. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=57127" rel="nofollow">A brief Introduction to Boosting</a> </td> 
 <td> 还是boosting </td> 
 </tr> 
 <tr> 
 <td> D M Gavrila. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=31883" rel="nofollow">The visual analysis of human movements: a survey</a> </td> 
 <td> 综述文章的引用一般都比较高 </td> 
 </tr> 
 <tr> 
 <td> Y Rui, T S Huang, S F Change. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=61128" rel="nofollow">Image retrieval: current techniques, promising directions, and open issues</a> </td> 
 <td> TSHuang小组对检索的一个总结 </td> 
 </tr> 
 <tr> 
 <td> J K Aggarwal, Q Cai. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=23570" rel="nofollow">Human motion analysis: a review</a> </td> 
 <td> 人体运动分析的一个综述 </td> 
 </tr> 
 </tbody> 
</table>

# 2000年 #

世纪之交，各种综述都出来了

<table> 
 <tbody> 
 <tr> 
 <td> J Shi, J Malik. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=7371" rel="nofollow">Normalized Cuts and Image Segmentation</a> </td> 
 <td> NCuts的引用率相当高，Jianbo Shi也因为这篇文章成为计算机视觉界引用率最高的作者之一 </td> 
 </tr> 
 <tr> 
 <td> Z Zhang. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=18030" rel="nofollow">A Flexible New Technique for Camera Calibration</a> </td> 
 <td> 张正友的关于摄像机标定的经典短文 </td> 
 </tr> 
 <tr> 
 <td> A K Jain, R P W Duin, J C Mao. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=66933" rel="nofollow">Statistical pattern recognition: a review</a>. </td> 
 <td> 统计模式识别综述，这一年pami上两篇很有名的综述之一。在这里推荐Web写的Statistical Pattern Recognition第三版，相当不错，网上有电子版。 </td> 
 </tr> 
 <tr> 
 <td> C Stauffe </td> 
 <td> Learning Patterns of Activity Using Real-Time Tracking </td> 
 <td> 搜TLD的时候发现这篇文章引用率也很高，两千多次。还没来得及读。 </td> 
 </tr> 
 <tr> 
 <td> D Taubman. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=20536" rel="nofollow">High performance Scalable Image Compression With EBCOT</a> </td> 
 <td> EBCOT，JPEG2000中的算法 </td> 
 </tr> 
 <tr> 
 <td> A W M Smeulders, M Worring, S Santini, A Gupta, R Jain. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=15607" rel="nofollow">Content-based image retrieval at the end of the early years</a> </td> 
 <td> 在世纪之交对图像检索的一篇很权威的综述。感觉在这之后检索的研究也没那么热了。不过在工业界热度依旧，各大网上购物平台，比如淘宝， 亚马逊，京东等都在做这方面的研发，衣服检索是一个很不错的应用点。 </td> 
 </tr> 
 <tr> 
 <td> M Pantic, L J M Rothkrantz. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=39394" rel="nofollow">Automatic analysis of facial expressions: the state of the art</a>. </td> 
 <td> &nbsp; </td> 
 </tr> 
 <tr> 
 <td> N Paragios, R Deriche. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=46527" rel="nofollow">Geodesic active contours and level sets for the detection and tracking of moving objects</a> </td> 
 <td> 使用level set做跟踪 </td> 
 </tr> 
 <tr> 
 <td> Y Rubner, C Tomasi, L Guibas. </td> 
 <td> T<a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=63175" rel="nofollow">The earth mover’s distance as a metric for image retrieval</a>. </td> 
 <td> EMD算法。Tomasi再次出现 </td> 
 </tr> 
 <tr> 
 <td> &nbsp; </td> 
 <td> PicToSeek Combining Color and Shape Invariant Features for Image Retrieval </td> 
 <td> 依然是检索特征 </td> 
 </tr> 
 </tbody> 
</table>

# 2001年 #

<table> 
 <tbody> 
 <tr> 
 <td> Paul Viola, Michael J Jones. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=14514" rel="nofollow">Robust real-time object detection</a> </td> 
 <td> 这是一篇很牛的文章，在人脸检测上几乎成了标准。比较坑爹的是，号称发在IJCV2001上，但怎么找也找不到。应该是IJCV2004年的那篇“Robust real-time face detection”吧。 他们在这一年另一篇比较出名的文章是在CVPR上的“Rapid Object Detection using a Boosted Cascade of Simple Features”这篇才是04年那篇著名文章的会议版。 </td> 
 </tr> 
 <tr> 
 <td> Y Boykov, Kolmogorov. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=31065" rel="nofollow">An experimental comparison of min-cut/max-flow algorithms for energy minimization in vision</a>. </td> 
 <td> 俄罗斯人在graph cut领域开始发力了 </td> 
 </tr> 
 <tr> 
 <td> T Moeslund, E Granum. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=48260" rel="nofollow">A Survey of Computer Vision Based Human Motion Capture</a> </td> 
 <td> 人体运动综述 </td> 
 </tr> 
 <tr> 
 <td> T F Chan, L Vese. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=94020" rel="nofollow">Active contours without edges</a>. </td> 
 <td> Snake和level set领域的经典文章 </td> 
 </tr> 
 <tr> 
 <td> A M Martinez, A C Kak. </td> 
 <td> <a href="http://citeseer.ist.psu.edu/viewdoc/summary?cid=132961" rel="nofollow">PCA versus LDA</a> </td> 
 <td> PCA也是计算机视觉中非掌握不可的工具。LDA在模式识别中有很重要的地位 </td> 
 </tr> 
 <tr> 
 <td> BS Manjunath </td> 
 <td> <a href="http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=927424" rel="nofollow">Color and texture descriptors</a> &nbsp; </td> 
 <td> 颜色和纹理的描述子，在识别中很有用 </td> 
 </tr> 
 </tbody> 
</table>

# 2002年 #

<table> 
 <tbody> 
 <tr> 
 <td> D Comaniciu, P Meer. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=7926" rel="nofollow">Mean shift: A robust approach toward feature space analysis</a>.&nbsp; </td> 
 <td> Mean shift的经典文章。前两天发现Comaniciu已经是IEEE Fellow了 </td> 
 </tr> 
 <tr> 
 <td> Ming-Husan Yang, David J Kriegman, Narendra Ahuja. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=32130" rel="nofollow">Detecting Faces in Images: A Survey</a>. </td> 
 <td> 人脸检测综述，引用率想不高都难 </td> 
 </tr> 
 <tr> 
 <td> R Hsu, M Abdel-Mottaleb. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=66056" rel="nofollow">Face Detection in Color Images</a>. </td> 
 <td> 依然是人脸检测，名字都起得这么霸气 </td> 
 </tr> 
 <tr> 
 <td> J-L Starck, E J Candès, D L Donoho. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=5282" rel="nofollow">The curvelet transform for image denoising</a>. </td> 
 <td> Geometrical wavelet中的一篇代表作。其他的如ridgelet, contourlet, bandelet等在这里就不赘述了。研究这方面的很容易找到这方面的经典文献。个人以为不研究这方面的看了后对自己的研究也不会有多大启发。曾经以为这个方向会很火，到最后还是没火起来。 我觉得原因可能是现在存储和传输能力的大大提高，使得对压缩的需求没有那么大了，这方面的研究自然就停滞了，就如同JPEG2000没有成气候一样。 </td> 
 </tr> 
 <tr> 
 <td> &nbsp; </td> 
 <td> Shape matching and object recognition using shape contexts </td> 
 <td> Shape context。用形状匹配达到目标识别目的。这方面最经典的文章了。随后后续也有一些这方面的文章，但基本都是很小的改进或者应用。作者提供了原码，可以在matlab上运行看看效果。 </td> 
 </tr> 
 <tr> 
 <td> N Paragios, R Deriche. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=46530" rel="nofollow">Geodesic active regions and level set methods for supervised texture segmentation</a> </td> 
 <td> &nbsp; </td> 
 </tr> 
 <tr> 
 <td> &nbsp; </td> 
 <td> <a href="http://www.cc.gatech.edu/%7Erehg/Papers/SkinDetect-IJCV.pdf" rel="nofollow">Statistical Color Models with Application to Skin Detection</a> </td> 
 <td> &nbsp; </td> 
 </tr> 
 <tr> 
 <td> &nbsp; </td> 
 <td> A tutorial on particle filters for online nonlinear non-Gaussian Bayesian tracking </td> 
 <td> particle filter的一个综述 </td> 
 </tr> 
 </tbody> 
</table>

# 2003年 #

<table> 
 <tbody> 
 <tr> 
 <td> W Zhao, R Chellappa, P J Phillips, A Rosenfeld. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=85653" rel="nofollow">Face recognition: A literature survey</a>. </td> 
 <td> 人脸检测的综述 </td> 
 </tr> 
 <tr> 
 <td> J Sivic, A Zisserman. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=235322" rel="nofollow">Video Google: A text retrieval approach to object matching in videos</a>. I </td> 
 <td> 好像是Visual words的起源文章。引用率很高，先列出来再看。 </td> 
 </tr> 
 <tr> 
 <td> D Comaniciu, V Ramesch, P Meer. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=46525" rel="nofollow">Kernel-Based Object Tracking</a>. </td> 
 <td> 基于核的跟踪。 </td> 
 </tr> 
 <tr> 
 <td> B Zitová, J Flusser. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=241789" rel="nofollow">Image registration methods: A survey</a>. </td> 
 <td> 又一篇图像配准的综述。做图像配准的比较有福气，综述很多 </td> 
 </tr> 
 <tr> 
 <td> K Mikolajczyk, C Schmid. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=144094" rel="nofollow">A performance evaluation of local descriptors</a>.&nbsp; </td> 
 <td> 比较各种描述子的，包括SIFT </td> 
 </tr> 
 <tr> 
 <td> M J Wainwright, M I Jordan. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=265423" rel="nofollow">Graphical models, exponential families, and variational inference</a>.&nbsp; </td> 
 <td> 乔丹的名气太大，不露露脸说不过去 </td> 
 </tr> 
 <tr> 
 <td> J Portilla, V Strela, M Wainwright, E Simoncelli. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=408188" rel="nofollow">Image denoising using scale mixtures of gaussians in the wavelet domain</a>. </td> 
 <td> 图像去噪，小波变换，混合高斯 </td> 
 </tr> 
 <tr> 
 <td> Robert E. Schapire </td> 
 <td> The Boosting Approach to Machine Learning&nbsp; An Overview </td> 
 <td> boosting作者自己写的综述，自然值得一看。 </td> 
 </tr> 
 </tbody> 
</table>

# 2004年 #

<table> 
 <tbody> 
 <tr> 
 <td> &nbsp; </td> 
 <td> Lucas-Kanade 20 Years On A Unifying Framework </td> 
 <td> 引用文章摘要的第一句话 Since the Lucas-Kanade algorithm was proposed in 1981 image alignment has become one of the most widely used techniques in computer vision. Applications range from optical flow and tracking to layered motion, mosaic construction, and face coding. </td> 
 </tr> 
 <tr> 
 <td> D G Lowe. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=37223" rel="nofollow">Distinctive image features from scale-invariant keypoints</a>.&nbsp; </td> 
 <td> SIFT，不解释 </td> 
 </tr> 
 <tr> 
 <td> Chih-Chung Chang, Chih-Jen Lin. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=34919" rel="nofollow">LIBSVM: A library for support vector machines</a> </td> 
 <td> 我实在怀疑引用这篇文章的人是否都看过这篇文章。貌似不看这篇文章也可以使用LIBSVM </td> 
 </tr> 
 <tr> 
 <td> Z Wang, A C Bovik, H R Sheikh, E P Simoncelli. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=385167" rel="nofollow">Image quality assessment: From error visibility to structural similarity</a> </td> 
 <td> 图像质量评价，最近Bovik还有一篇类似的文章也刊登在TIP上 </td> 
 </tr> 
 <tr> 
 <td> Y Ke, R Sukthankar. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=144090" rel="nofollow">Pca-sift: a more distinctive representation for local image descriptors</a> </td> 
 <td> SIFT的变形 </td> 
 </tr> 
 <tr> 
 <td> &nbsp; </td> 
 <td> Review of shape representation and description techniques </td> 
 <td> &nbsp; </td> 
 </tr> 
 <tr> 
 <td> &nbsp; </td> 
 <td> Efficient Graph-Based Image Segmentation </td> 
 <td> &nbsp; </td> 
 </tr> 
 </tbody> 
</table>

# 2005年 #

<table> 
 <tbody> 
 <tr> 
 <td> N Dalal, B Triggs. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=296330" rel="nofollow">Histograms of oriented gradients for human detection</a>.&nbsp; </td> 
 <td> HOG 虽然很新，但很经典 </td> 
 </tr> 
 <tr> 
 <td> A C Berg, T L Berg, J Malik. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=247179" rel="nofollow">Shape matching and object recognition using low distortion correspondences</a>. </td> 
 <td> 还是shape matching </td> 
 </tr> 
 <tr> 
 <td> S Roth, M Black. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=385165" rel="nofollow">Fields of experts: A framework for learning image priors</a>. </td> 
 <td> 这篇应该要归结到图像统计特性的范畴吧 </td> 
 </tr> 
 <tr> 
 <td> Z Tu, X Chen, A L Yuille, S C Zhu. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=446316" rel="nofollow">Image parsing: Unifying segmentation, detection, and recognition</a>. </td> 
 <td> &nbsp; </td> 
 </tr> 
 <tr> 
 <td> &nbsp; </td> 
 <td> Geodesic active regions and level set methods for motion estimation and tracking </td> 
 <td> &nbsp; </td> 
 </tr> 
 <tr> 
 <td> Chunming Li, Chenyang Xu, Changfeng Gui, and&nbsp; Martin D. Fox </td> 
 <td> Level Set Evolution Without Re-initialization: A New Variational Formulation </td> 
 <td> 这篇文章解决了level set中需要不停的重初始化的问题。在2010年的TIP上有一篇Journal版本 <a href="http://www.engr.uconn.edu/%7Ecmli/paper/DRLSE.pdf" rel="nofollow">Distance Regularized Level Set Evolution and its Application to Image Segmentation</a> </td> 
 </tr> 
 <tr> 
 <td> &nbsp; </td> 
 <td> A Performance Evaluation of Local Descriptors </td> 
 <td> 前面那篇是会议的，这篇是PAMI上的。比较各种描述子的，包括SIFT </td> 
 </tr> 
 </tbody> 
</table>

# 2006年 #

<table> 
 <tbody> 
 <tr> 
 <td> D Donoho. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=309660" rel="nofollow">Compressed sensing</a>.&nbsp; </td> 
 <td> CS 压缩感知 最近很火的一个名词 </td> 
 </tr> 
 <tr> 
 <td> Greg Welch, Gary Bishop. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=19218" rel="nofollow">An introduction to the Kalman Filter</a>. </td> 
 <td> kalman滤波 </td> 
 </tr> 
 <tr> 
 <td> S Lazebnik, C Schmid, J Ponce. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=296140" rel="nofollow">Beyond bags of features: spatial pyramid matching for recognizing natural scene categories</a>.&nbsp; </td> 
 <td> Visual words </td> 
 </tr> 
 <tr> 
 <td> Xiaojin Zhu. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=347714" rel="nofollow">Semi-supervised learning literature survey</a>. </td> 
 <td> &nbsp; </td> 
 </tr> 
 <tr> 
 <td> A Yilmaz, O Javed, M Shah. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=1009790" rel="nofollow">Object Tracking: A survey</a>. </td> 
 <td> tracking的一篇综述 </td> 
 </tr> 
 <tr> 
 <td> &nbsp; </td> 
 <td> Image Alignment and Stitching: A Tutorial </td> 
 <td> &nbsp; </td> 
 </tr> 
 </tbody> 
</table>

# 2007年 #

<table> 
 <tbody> 
 <tr> 
 <td> &nbsp; </td> 
 <td> A Review of Statistical Approaches to Level Set Segmentation: Integrating Color, Texture, Motion and Shape </td> 
 <td> &nbsp; </td> 
 </tr> 
 <tr> 
 <td> &nbsp; </td> 
 <td> The Appearance of Human Skin: A Survey </td> 
 <td> &nbsp; </td> 
 </tr> 
 <tr> 
 <td> &nbsp; </td> 
 <td> Local Invariant Feature Detectors: A Survey </td> 
 <td> &nbsp; </td> 
 </tr> 
 </tbody> 
</table>

# 2008年 #

<table> 
 <tbody> 
 <tr> 
 <td> H Bay, A Ess, T Tuytelaars, L V Gool. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=440048" rel="nofollow">SURF: Speeded Up Robust Features</a>.&nbsp; </td> 
 <td> &nbsp; </td> 
 </tr> 
 <tr> 
 <td> K E A van de Sande, T Gevers, C G M Snoek. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=9553947" rel="nofollow">Evaluation of Color Descriptors for Object and Scene Recognition</a> </td> 
 <td> &nbsp; </td> 
 </tr> 
 <tr> 
 <td> M Yang&nbsp; </td> 
 <td> <a href="http://hal.archives-ouvertes.fr/docs/00/44/60/37/PDF/ARS-Journal-SurveyPatternRecognition.pdf" rel="nofollow">A Survey of Shape Feature Extraction Techniques</a> &nbsp; </td> 
 <td> 虽然这篇文章的引用率目前来看并不高,但个人认为这是一篇在shape feature方面很不错的文章 </td> 
 </tr> 
 <tr> 
 <td> P. Felzenszwalb, D. McAllester, D. Ramanan </td> 
 <td> A Discriminatively Trained, Multiscale, Deformable Part Model </td> 
 <td> 2008年的CVPR，到现在引用已有四百多次，潜力巨大。rosepink提供 </td> 
 </tr> 
 </tbody> 
</table>

# 2009年 #

<table> 
 <tbody> 
 <tr> 
 <td> J Wright, A Y Yang, A Ganesh, S S Sastry, Ma. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=5118398" rel="nofollow">&nbsp;Robust Face Recognition via Sparse Representation</a>.&nbsp; </td> 
 <td> &nbsp; </td> 
 </tr> 
 <tr> 
 <td> B Settles. </td> 
 <td> <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?cid=10168311" rel="nofollow">Active learning literature survey</a> </td> 
 <td> &nbsp; </td> 
 </tr> 
 </tbody> 
</table>

# 2010年 #

# 2011年 #

<table> 
 <tbody> 
 <tr> 
 <td> &nbsp; </td> 
 <td> Hough Forests for Object Detection, Tracking, and Action Recognition </td> 
 <td> &nbsp; </td> 
 </tr> 
 <tr> 
 <td> &nbsp; </td> 
 <td> Robust Principal Component Analysis? </td> 
 <td> Candes 和UIUC的Ma Yi等人 </td> 
 </tr> 
 </tbody> 
</table>

# 2012年 #

<table> 
 <tbody> 
 <tr> 
 <td> Zdenek Kalal, Krystian Mikolajczyk, and Jiri Matas, </td> 
 <td> Tracking-Learning-Detection </td> 
 <td> PAMI上的，虽然还没有正式发表，但肯定会火。在作者的主页上有几篇相关的会议文章，demo和code。用到了Lucas-Kanade方法 </td> 
 </tr> 
 <tr> 
 <td> &nbsp; </td> 
 <td> &nbsp; </td> 
 <td> &nbsp; </td> 
 </tr> 
 </tbody> 
</table>

[http_iask.sina.com.cn_u_2252291285_ish_folderid_775855]: http://iask.sina.com.cn/u/2252291285/ish?folderid=775855