目前已经有大量的特征提取方法,如何选择其中最具代表性的信息是一件非常头疼的事情。Allen Y. Yang等人从利用稀疏化表示方法介绍了一种新的特征选择的方法。这种特征选择策略不需要任何花哨的空间变换方法或者复杂的分类器设计。只需要将测试对象映射到训练样本点空间组成的冗余原子库空间中,查看得到的稀疏化表示的能量集中在哪类样本组成的冗余原子库中,由此推断测试样本所属的类别。 有几个问题值得考虑: 1 什么样的稀疏表示适合于这种识别方法,L 1 ,L 2 或者 L p ? 2 需要多少样本来完成这样的识别任务,即原子库规模的问题;感觉上,样本比较多的时候识别效果可能会比较好,但是否存在极限情况,即样本越多越好?
2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2010) 会议网址: http:// cvl. umiacs. umd. edu/ conferences/ cvpr2010/ 论文摘要提交截止日期:2009年11月12日 会议召开时间地点:San Francisco, California, USA, June 13, 2010-June 18, 2010 这是人工智能与相关学科领域的顶级会议,2010年已经是27届,每年会议均被EI、ISTP收录,每年的会议论文会选择部分论文到SCI收录期刊发表。2008年26届会议收录756篇论文,其中美国361篇、中国63篇、德国44篇、法国25篇、英国24篇、加拿大23篇、日本22篇、 澳大利亚 19篇、以色列16篇,新加坡16篇、瑞士16篇、新西兰14篇、西班牙14篇等。 中科院12篇、香港中文大学9篇、清华7篇、中科大6篇、浙大5篇、香港科技大学5篇等著名学校均在26届会议上发表论文。 Topics : Motion and Tracking Stereo and Structure from Motion Shape-from-X Color and Texture Segmentation and Grouping Image-Based Modeling Illumination and Reflectance Modeling Shape Representation and Matching Sensors Early and Biologically-Inspired Vision Computational Photography and Video Object Recognition Object Detection and Categorization Video Analysis and Event Recognition Face and Gesture Analysis Statistical Methods and Learning Performance Evaluation Medical Image Analysis Image and Video Retrieval Vision for Graphics Vision for Robotics Applications of Computer Vision CVPR 2010
会议网址: http://www.icpr2010.org/ 论文提交截止日期:2010年1月15号 Istanbul Convention Exhibition Centre (ICEC), Istanbul, Turkey August 23-26, 2010 这是国际模式识别学会( International Association for Pattern Recognition (IAPR) .)的会议,每两年召开一次,每年会议均被EI、ISTP收录,19届会议收录987篇论文,其中美国203篇、中国181篇、日本97篇、法国68篇、德国46篇、加拿大35篇。 中科院、清华、哈工大、上海交大、北大、香港城市大学、香港理工大学等著名学校均在19届会议上发表论文。 Topics Track I: Computer Vision * Segmentation and Grouping * Color and Texture * Shape Modeling * Object Detection and Recognition * Stereo and Motion * Scene Understanding * Image based modeling * Tracking and Surveillance * Performance Evaluation * Biologically Motivated vision * Applications Track II: Pattern Recognition and Machine Learning * Statistical machine learning * Structural and Syntactic Pattern Recognition * Neural Networks * Kernel Methods * Graphical Models * Bayesian Methods * Combinatorial Optimization * Feature Selection * Model Selection * Data Mining and Pattern Recognition for Bioinformatics * Voting and Fusion Schemes * Compressed Sensing Detection and Estimation * Applications Track III: Signal, Speech, Image and Video Processing * Speech Recognition * Speech Processing and Synthesis * Image and Multidimensional Signal Processing * Image and Video Synthesis and Geometric Compression * Motion Estimation and Video Registration * Video Processing and Motion-Compensated Filtering * Video Compression and Streaming * Multimedia Signal Processing * Multimedia Security and Perceptual Hashes * Signal Processing for Sensor Networks * Applications Track IV: Biometrics and Human Computer Interaction * Fingerprint Recognition * Face Recognition * Iris Recognition * Speaker Recognition * Palmprint, Gait and Other Biometrics * Multimodal Person Verification and Identification * Biometric Security and Privacy * Facial Expression Analysis * Gesture Analysis * Applications Track V: Multimedia and Document Analysis, Processing and Retrieval * Digital Libraries * Multimedia Content Description and Indexing * Retrieval and Browsing * Multi-modal Interaction and Integration * Summarization * Data Mining and Machine Learning for Multimedia Information Retrieval * Video analysis and event recognition * 3D Video Processing and Compression * Applications Track VI: Medical Imaging and Visualization * General Medical Image Computing * Quantitative Medical Image Analysis * Computer Aided Detection and Diagnosis * Molecular and Cellular Image Analysis * Interventional Systems and Medical Robotics * Computational Neuroimaging * General Biological Image Computing * Visualization and Interaction * Applications