Robustnesstochangesinilluminationconditionsaswellas viewing perspectives is an important requirement formany computer vision applications. One of the key fac-ors in enhancing the robustness of dynamic scene analy-sis that of accurate and reliable means for shadow de-ection. Shadowdetectioniscriticalforcorrectobjectde-ection in image sequences. Many algorithms have beenproposed in the literature that deal with shadows. How-ever,acomparativeevaluationoftheexistingapproachesisstill lacking. In this paper, the full range of problems un-derlyingtheshadowdetectionareidenti?edanddiscussed.Weclassifytheproposedsolutionstothisproblemusingaaxonomyoffourmainclasses, calleddeterministicmodeland non-model based and statistical parametric and non-parametric. Novelquantitative(detectionanddiscrimina-ionaccuracy)andqualitativemetrics(sceneandobjectin-dependence,?exibilitytoshadowsituationsandrobustnesso noise) are proposed to evaluate these classes of algo-rithms on a benchmark suite of indoor and outdoor videosequences.
標簽： Robustnesstochangesinillumination conditionsaswellas perspectives requirement
上傳時間： 2014-01-23
上傳用戶：whenfly
來自澳大利亞Qeensland大學的計算機視覺Matlab工具箱。 This Toolbox provides a number of functions that are useful in computer vision, machine vision and related areas. It is a somewhat eclectic collection reflecting the author s personal interest in areas of photometry, photogrammetry, colorimetry. It covers functions such as image file reading and writing, filtering, segmentation, feature extraction, camera calibration, camera exterior orientation, display, color space conversion and blackbody radiators. The Toolbox, combined with MATLAB and a modern workstation computer, is a useful and convenient environment for investigation of machine vision algorithms. It is possible to use MEX files to interface with image acquisition hardware ranging from simple framegrabbers to Datacube servers.
標簽： Qeensland functions provides Toolbox
上傳時間： 2015-09-30
上傳用戶：qb1993225
計算機視覺，COMPUTER VISION， 國外的課件，希望對大家有用
標簽： 計算機視覺
上傳時間： 2015-10-03
上傳用戶：wang5829
This handbook presents a thorough overview in 45 chapters from more than 100 renowned experts in the field. It provides the tools to help overcome the problems of video storage, cataloging, and retrieval, by exploring content standardization and other content classification and analysis methods. The challenge of these complex problems make this book a must-have for video database practitioners in the fields of image and video processing, computer vision, multimedia systems, data mining, and many other diverse disciplines. Topics include video segmentation and summarization, archiving and retrieval, and modeling and representation.
標簽： handbook chapters presents overview
上傳時間： 2013-12-15
上傳用戶：yuzsu
penMesh is a generic and efficient data structure for representing and manipulating polygonal meshes. OpenMesh is developed at the Computer Graphics Group, RWTH Aachen , as part of the OpenSGPlus project, is funded by the German Ministry for Research and Education ( BMBF), and will serve as geometry kernel upon which the so-called high level primitives (e.g. subdivision surfaces or progressive meshes) of OpenSGPlus are built. It was designed with the following goals in mind : Flexibility : provide a basis for many different algorithms without the need for adaptation. Efficiency : maximize time efficiency while keeping memory usage as low as possible. Ease of use : wrap complex internal structure in an easy-to-use interface.
標簽： manipulating representing and efficient
上傳時間： 2015-10-14
上傳用戶：米卡
Learning Kernel Classifiers: Theory and Algorithms, Introduction This chapter introduces the general problem of machine learning and how it relates to statistical inference. 1.1 The Learning Problem and (Statistical) Inference It was only a few years after the introduction of the first computer that one of man’s greatest dreams seemed to be realizable—artificial intelligence. Bearing in mind that in the early days the most powerful computers had much less computational power than a cell phone today, it comes as no surprise that much theoretical research on the potential of machines’ capabilities to learn took place at this time. This becomes a computational problem as soon as the dataset gets larger than a few hundred examples.
標簽： Introduction Classifiers Algorithms introduces
上傳時間： 2015-10-20
上傳用戶：aeiouetla
Some algorithms of variable step size LMS adaptive filtering are studied．The VS—LMS algorithm is improved． Another new non-linear function between肛and e(／ t)is established．The theoretic analysis and computer simulation results show that this algorithm converges more quickly than the origina1．Furthermore，better antinoise property is exhibited under Low—SNR environment than the original one．
標簽： algorithms LMS algorithm filtering
上傳時間： 2014-01-23
上傳用戶：yxgi5
his paper provides a tutorial and survey of methods for parameterizing surfaces with a view to applications in geometric modelling and computer graphics. We gather various concepts from di® erential geometry which are relevant to surface mapping and use them to understand the strengths and weaknesses of the many methods for parameterizing piecewise linear surfaces and their relationship to one another.
標簽： parameterizing provides tutorial surfaces
上傳時間： 2014-11-09
上傳用戶：努力努力再努力
Verilog HDL: Magnitude For a vector (a,b), the magnitude representation is the following: A common approach to implementing these arithmetic functions is to use the Coordinate Rotation Digital Computer (CORDIC) algorithm. The CORDIC algorithm calculates the trigonometric functions of sine, cosine, magnitude, and phase using an iterative process. It is made up of a series of micro-rotations of the vector by a set of predetermined constants, which are powers of two. Using binary arithmetic, this algorithm essentially replaces multipliers with shift and add operations. In a Stratix™ device, it is possible to calculate some of these arithmetic functions directly, without having to implement the CORDIC algorithm.
標簽： representation Magnitude the magnitude
上傳時間： 2013-12-24
上傳用戶：金宜
Web technology is not evolving in comfortable and incremental steps, but i s turbulent, erratic, and often rather uncomfortable. It is estimated that the Internet, arguably the most important part of the new technological environment, has expanded by about 2000 ％ and that is doubling in size every six to ten months. In recent years, the advance in computer and web technologies and the decrease in their cost have expanded the means available to collect and store data. As an intermediate consequence, the amount of information (Meaningful data) stored has been increasing at a very fast pace.
標簽： comfortable incremental technology and
上傳時間： 2015-11-05
上傳用戶：Shaikh
