tag:blogger.com,1999:blog-35439708032496715242012-02-17T08:38:47.029+08:00astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.comBlogger251100tag:blogger.com,1999:blog-3543970803249671524.post-61382607334597862172009-07-24T11:02:00.003+08:002009-07-24T11:33:59.090+08:00

"Hands-free vision-based interface forcomputer accessibility", Javier Varona, Cristina Manresa-Yee, Francisco J. Perales, JNCA08'In order to draw disabled people to new technology, the paper presents a hands-free vision-based interface. First, tracking the facial feature in few frames to initialize the model. No special lighting or static background is important, any orientation must be avoided

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com1tag:blogger.com,1999:blog-3543970803249671524.post-10643839484564206472009-06-18T20:33:00.001+08:002009-07-24T09:59:25.931+08:00

"Support vector learning for ordinal regression," R. Herbrich, ICANN, 1999The paper presents a method to solve ordinal regression by support vector. They reformulate the ranking problem into binary classification problem. That is, given the pairs of instances, output their relative ranking according to their classified labels. The idea about applying SVM on ranking is really impressive. However,

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com0tag:blogger.com,1999:blog-3543970803249671524.post-549774806343934392009-06-03T00:06:00.002+08:002009-07-24T09:59:13.651+08:00

"The structure and function of complex networks," Newman, 2003.Generally, there are some terms with regard to a graph, such as vertex, edge, directed/undirected, degree, component, geodesic path, and diameter. Besides, there are also some kinds of networks in the real world. For example, social networks is groups of people with some pattern of interactions between them. The information networks

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com0tag:blogger.com,1999:blog-3543970803249671524.post-62454092887150275832009-05-27T03:18:00.002+08:002009-07-24T09:58:48.910+08:00

"Lazy Snapping," Li, et. al., ACM SIGGRAPH, 2004The paper presents a coarse-to-fine UI design for image cutout which is the technique of removing an object in an image from the background. There are two steps : (1) object marking (2) boundary editing. In the first step, user can specify the foreground and background region respectively. Then, the uncertain region can be labeled by calculating the

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com0tag:blogger.com,1999:blog-3543970803249671524.post-31951665619966003042009-05-20T01:33:00.002+08:002009-07-24T09:58:23.443+08:00

"Learning Low-Level Vision," Freeman, IJCV, 2000.The paper describes a learning-based method for low-level vision problems, such as motion analysis, inferring shape and reflectance from a photograph, or extrapolating image detail, that is, how to estimate scenes from images is the goal. Given training sets, they succeed in enumerating a coarse sampling of all input patch values by preprocessing

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com0tag:blogger.com,1999:blog-3543970803249671524.post-57284476140558107222009-05-20T01:13:00.004+08:002009-07-24T09:58:14.879+08:00

"An introduction to graphical models," Kevin Murphy, 2001.Graphical models are the combination of probability theory and graph theory. They can be directed or undirected models. The directed graphical models are known as Bayesian networks, where the ancestor/parent relationship is with respect to some fixed topological ordering of the nodes, and the undirected graphical models are known as Markov

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com0tag:blogger.com,1999:blog-3543970803249671524.post-91921807985949084982009-05-13T11:31:00.002+08:002009-07-24T09:57:58.546+08:00

"Rapid object detection using a boosted cascade of simple features," Paul Viola and Michael Jones, CVPR, 2001The paper proposes a rapid object detection method and constructs a frontal face detection system to prove the idea. First, they brings the “integral image” calculation which lets feature evaluation faster. That is, do first pass to sum the region of pixels above and to the left of each

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com0tag:blogger.com,1999:blog-3543970803249671524.post-48821205081381573422009-05-07T01:50:00.002+08:002009-07-24T09:57:10.027+08:00

"Names and Faces in the News Abstract," Miller, CVPR, 2004The paper presents a clustering method to deal with ambiguities in labeling and identify incorrectly labeled faces. Before clustering, they perform kernel PCA to reduce the dimensionality and linear discriminant analysis to project data for discriminant analysis.The clustering process goes as following: 1.Randomly assign each image to one

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com0tag:blogger.com,1999:blog-3543970803249671524.post-7599514128221666812009-05-06T20:32:00.002+08:002009-07-24T09:56:53.134+08:00

"Object Recognition as Machine Translation: Learning a Lexicon for a Fixed Image Vocabulary," Duygulu, ECCV, 2002In order to learn a lexicon for a fixed image vocabulary, there are three major object recognition problem which should be discussed : 1. What counts as an object? All words.2. Which objects are easy to recognize? The words attached to image regions.3. Which objects are

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com0tag:blogger.com,1999:blog-3543970803249671524.post-80312704506236494182009-05-06T13:58:00.002+08:002009-07-24T09:56:31.540+08:00

"On Spectral Clustering: Analysis and an algorithm", Andrew Y. Ng, Michael I. Jordan, Yair Weiss, NIPS 2001Based on the spectral methods which use the top eigenvectors of a matrix corresponding to the similarity between some features, the paper proposes a simple spectral clustering algorithm to utilize the k eigenvectors simultaneously. There are several steps:1.Construct an affinity matrix

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com1tag:blogger.com,1999:blog-3543970803249671524.post-31684021031464240752009-05-06T12:01:00.001+08:002009-07-24T09:56:19.705+08:00

"Normalized Cuts and Image Segmentation", Jianbo Shi and Jitendra Malik, Trans. PAMI 2000The paper proposes a general framework for image segmentation. Generally, the result of each partition method is affected by the coherence of brightness, color, texture, or motion, and the hierarchical partition should form a tree structure. Therefore, the authors generate a graph theoretic formulation of

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com0tag:blogger.com,1999:blog-3543970803249671524.post-16936626343714469322009-04-07T23:58:00.001+08:002009-07-24T09:55:52.139+08:00

"Algorithms for fast vector quantization," Sunil. Arya and David. M. Mount, 1993The paper presents three algorithms, standard k-d tree search, priority k-d tree search, and neighborhood graph search, for nearest neighbor matching problem which is important in many applications, including vector quantization. Standard k-d tree is a binary search tree in high dimension. Each internal node of the

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com0tag:blogger.com,1999:blog-3543970803249671524.post-67281352137651013672009-03-31T23:56:00.005+08:002009-07-24T09:55:24.253+08:00

"Latent Dirichlet allocation," D. Blei, A. Ng, and M. Jordan. . Journal of Machine Learning Research, 3:993–1022, January 2003Latent Dirichlet allocation (LDA) is a generative probabilistic model for collections of discrete data such as text corpora. The goal is to find short descriptions of the members of a collection that enable efficient processing of large collections while preserving the

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com0tag:blogger.com,1999:blog-3543970803249671524.post-34973370653337620632009-03-31T22:54:00.002+08:002009-07-24T09:55:12.865+08:00

“Probabilistic Latent Semantic Indexing”, Thomas Hofmann, SIGIR, 1999 Probabilistic Latent Semantic Analysis(PLSA) is a novel method for automated indexing based on the likelihood principle. PLSA defines a statistical latent class model called aspect model fitting with tempered EM algorithm. In contrast to LSA which determines the optimal decomposition by L2-norm, PLSA relies on the likelihood

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com0tag:blogger.com,1999:blog-3543970803249671524.post-75433615721923241582009-03-27T14:36:00.002+08:002009-07-24T09:54:33.465+08:00

Title: Contour and Texture Analysis for Image SegmentationAuthor: JITENDRA MALIK, SERGE BELONGIE, THOMAS LEUNG∗AND JIANBO SHI†Publisher: IJCVDate of Publication: Feb 23, 2001Because texture provides a good local descriptor of image patches, this paper proposes a general algorithm for image segmentation by using texture. They apply different filters to the input images and get the filter responses

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com0tag:blogger.com,1999:blog-3543970803249671524.post-40837486041865616422009-03-22T12:57:00.001+08:002009-07-24T09:54:22.459+08:00

Title: Shape Matching and Object Recognition Using Shape ContextsAuthor: Serge Belongie, Jitendra Malik, Jan PuzinchaPublisher: IEEEMonth of Publication: April 2002 The paper proposes a stable and simple algorithm for finding corresponding between shapes. They introduce a shape descriptor, shape context, and maximize the similarity in the bipartite graph. The demonstration of 2D objects, e.g.,

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com0tag:blogger.com,1999:blog-3543970803249671524.post-73446803283497206502009-03-13T22:06:00.002+08:002009-07-24T09:54:07.001+08:00

Title : Nonlinear Dimensionality Reduction by Locally Linear EmbeddingAuthor : S. T. Roweis and L. K. SaulYear of Publication: 2000 The paper introduces an unsupervised learning algorithm, locally linear embedding(LLE). The following is the steps of locally linear embedding.1.Find the neighbors of each point which close to a locally linear patch the manifold. (kNN in this paper)2.Reconstruct

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com0tag:blogger.com,1999:blog-3543970803249671524.post-68339844354961136122009-03-09T00:40:00.002+08:002009-07-24T09:53:47.558+08:00

Title: Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear ProjectionAuthor: Peter N. Belhumeur, Joao P. Hespanha, and David J. kriegmanYear of Publication: 1997Publisher: IEEE In this paper, the authors proposed the method, Fisherfaces, for face recognition. Fisherfaces is insensitive to large variations in illumination and facial expressions. Compared with eigenfaces which

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com0tag:blogger.com,1999:blog-3543970803249671524.post-47565712077771465752009-03-08T21:08:00.003+08:002009-07-24T09:53:35.790+08:00

Title: Eigenfaces for RecognitionAuthor: Matthew Turk and Alex PentlandYear of Publication: 1991 Eigenfaces is a approach that decomposes face images into a small set of characteristic feature images and is based on information theory. In other words, the approach extracts the information contained in a collection of face images and uses the variation of these images to encode and compare

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com0tag:blogger.com,1999:blog-3543970803249671524.post-4871194472395255022009-03-03T22:15:00.004+08:002009-07-24T09:53:09.651+08:00

Title:Scale & Affine Invariant Interest Point DetectorsAuthor: KRYSTIAN MIKOLAJCZYK AND CORDELIA SCHMIDDate of Publication: January 22, 2004The paper describes two approaches for scale and affine invariant interest point detection. The scale invariant interest point detector, Harris-Laplace detector, combines the Harris detector with automatic scale selection. The algorithm involves a multi-scale

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com0tag:blogger.com,1999:blog-3543970803249671524.post-76267596515278519272009-03-03T21:29:00.004+08:002009-07-24T09:52:53.428+08:00

Title:Distinctive Image Featuresfrom Scale-Invariant KeypointsAuthor:David G. LoweDate of Publication: January 5, 2004This paper describes an approach, Scale Invariant Feature Transform(SIFT), which transforms image data into scale-invariant coordinates relative to local features. There are several major stages of computation.(1) The scale-space extrema detection stage searches all scales and

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com0tag:blogger.com,1999:blog-3543970803249671524.post-27948527272828222282009-03-03T00:03:00.004+08:002009-03-03T00:11:36.733+08:00

Title: Image Retrieval: Ideas, Influences, and Trends of the New AgeAuthors: RITENDRA DATTA, DHIRAJ JOSHI, JIA LI, and JAMES Z. WANGYear of Publication: 2008Publisher: ACM Content-based image retrieval is an technology helps to organize digital picture archives by their visual content. There are two gaps, sensory gap and semantic gap, which define and motivate most of the related problems. The

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com0tag:blogger.com,1999:blog-3543970803249671524.post-91192715951875878792009-02-28T22:32:00.006+08:002009-02-28T22:43:01.503+08:00

Reading : "How to give a good research talk," Jones et. al. For a person who gives a 30-60 minutes talk, there are several suggestions to follow. First, prepare appropriate content according to background knowledge of the audience. Omit unnecessary contents and remain the things that can convince the listeners of the primary topic. It may be more persuasive with precisely motivating

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com0tag:blogger.com,1999:blog-3543970803249671524.post-51899387960143095962009-02-28T22:24:00.003+08:002009-02-28T22:43:17.798+08:00

Reading : "How to Read a Paper," Keshav, ACM SIGCOMM Computer Communication Review 2007 This writer shares an efficient method to read a paper. Because researchers may spent a lot of hours reading papers and it is terrible if wasting much effort in the process, he introduces the three-pass approach. First of all, skim through the target paper quickly. Read only titles, abstract,

astral-beamshttp://www.blogger.com/profile/01217607908794039437noreply@blogger.com0tag:blogger.com,1999:blog-3543970803249671524.post-65458182287805548542008-02-16T22:12:00.002+08:002009-02-25T16:02:08.112+08:00

test

星界光束http://www.blogger.com/profile/12208626454074503917noreply@blogger.com1