PPT-Image Feature Descriptors

Kenton McHenry PhD Research Scientist Raster Images 092 093 094 097 062 037 085 097 093 092 099 095 089 082 089 056 031 075 092 081 095 091 089 072. Categorization. . With. . Bags. of . Keypoints. Original . Authors. :. G.. . Csurka. , C.R. Dance, L. Fan, . J. . Willamowski. , C. Bray. ECCV Workshop on . Statistical. Learning in Computer – 2004. CSE P 576. Larry Zitnick (. larryz@microsoft.com. ). 20,000 images of Rome. =. ?. Large scale matching. How do we match millions or billions of images in under a second?. Is it even possible to store the information necessary?. . local. image . descriptors. . into. . compact. . codes. Authors. :. Hervé. . Jegou. Florent. . Perroonnin. Matthijs. . Douze. Jorge. . Sánchez. Patrick . Pérez. Cordelia. Schmidt. Presented. Segment Descriptor. Segments are areas of memory defined by a programmer and can be a code, data or stack segment.. In 80386 segments need not be all the same size and aligned. And segments need not be exactly 64 KB long, but we can define them to be any length from 1 byte to 4 GB.. Matthew Brown. University of British Columbia. (prev.) Microsoft Research. [ Collaborators: . †. Simon Winder, *Gang . Hua. , . †. Rick . Szeliski. . †. =MS Research, *=MS Live Labs]. Applications @MSFT. Detection: . introduction. Approaches. Holistic detection: use local search window that meets . criterias. Part-based detection: pedestrian as a collection of parts (to be found!). Patch-based detection: local features matched against a (learned) codebook, then voting for final detection. Eric Brenner. Paul Carpenter. Daniel Ehrenberg. Aaron McCarty. Travis Raines. Advised by Jeff . Ondich. Defining the Problem. What is optical character recognition (OCR)?. Input: an image of some text. 3 types of descriptors. :. SIFT / PCA-SIFT . (. Ke. , . Sukthankar. ). GLOH . (. Mikolajczyk. , . Schmid. ). DAISY . (. Tola. , et al, Winder, et al). Comparison of descriptors . (. Mikolajczyk. Klaus Mueller. Computer Science. Lab for Visual Analytics and Imaging (VAI). Stony Brook University. Wei Xu, Sungsoo Ha and Klaus Mueller. Motivation. Low-dose CT:. * Images from Google.com . Motivation. CS5670: Computer Vision. Noah Snavely. Reading. Szeliski: 4.1. Announcements. Project 1 Artifacts due tomorrow, Friday 2/17, at 11:59pm. Project 2 will be released next week. In-class quiz at the beginning of class Thursday. Principle Component Analysis. Why Dimensionality Reduction?. It becomes more difficult to extract meaningful conclusions from a data set as data dimensionality increases--------D. L. . Donoho. Curse of dimensionality. Samantha Horvath. Learning Based Methods in Vision. 2/14/2012. Introduction. Computer vision makes use of many “hand-crafted” descriptors.. These descriptors share many common components. This paper presents a modular framework for designing and optimizing new feature descriptors . the Classroom Environment . and Culture. . and . Professional . Collaboration . and . Communication . Dimensions of 5D . Norms for Learning. Talk. Listen. Share ideas. Respect opinions and ideas shared. Lycium. . barbarum. (goji) puree. Monica . Rosa . Loizzo. 1,*. , . Antonio . Mincione. 1. , . Rosa . Tundis. 1. , . Vincenzo . Sicari. 2. 1. Department . of . Pharmacy, . Health and . Nutritional .