BBYR Achieve
返回信息流
这是一条镜像帖。来源:北邮人论坛 / go-abroad / #120195同步于 2009/6/1
该镜像源已超过 30 天没有更新,可能在源站已被删除。
GoAbroad机器人发帖

【飞跃重洋】【领域介绍】Image/Video Processing Session,初

eeMars
2009/6/1镜像同步13 回复
为免以讹传讹,请大家多提提修改意见。 飞跃重洋 领域介绍 Image/Video Processing Session,初稿,征求意见 报名写写 各个领域介绍; 领域内的会议,牛导; Image/Video processing 匆忙之中,难免中英文混杂,请见谅; 1, Top Conference & Journal Top Conference 注1:有些是简写,Google就可以知道全称• 注2:这里只是相对TOP级别的,肯定有些TOP的我漏了,或者我觉得声誉挺好的,但是您觉得是水会的。仁者见仁。比如信号与系统那个牛会,我一直觉得不是很TOP,只是因为我有个朋友一次中两篇,但是发现认可度还是不错的。 注3:知名度大并不意味着一定难中; 注4:还有无穷多的大水会,非常容易中~不要局限在这个list里。 从知名度上看,SPIE, ICIP, ICME, ICIP, Visual System都是Top级的,但是都不是最难中的; 比较难中的有ACM的multimedia; SIGGRAPH最近也加入了Multimedia的session, 更难中一些。 如果时间有限,可以重点投知名度最高的几个会,其实不是特别难中,但是领域内的人应该都知道。说白了就是大会;这样起的效果是申请的时候让教授知道,行,这小伙英文书写没大问题,也能干活。 如果觉得自己Idea很牛,可以冲击后面两个难中的,可以向教授证明自己Smart。让美国人认为自己能干比较容易,证明Smart,是一个更高阶的过程。 Journal 如果本科,或者说是研究生,发Journal 时间可能比较紧张。列几个TOP的期刊,仅供参考: Transactions on Image Processing;顶级 CSVT,在视频处理,尤其是编码方向顶级; 此外,Multimedia Systems,IEEE Transactions on Multimedia也不错。 有两本相关,级别很高,但是不全是视频图像的: PAMI,Signal processing, Information theory 其中PAMI声誉很好。 2, Top Professors最TOP的? 牛人很多。完全有可能某个藏在角落的AP,我没听过,你也没听过,但是做个什么东西,一下就有名了。。。 我这里列的老板符合几个情况: 1,相对功成名就; 2,我知道; 3,中国人优先考虑~ 注意:不一定牛老板,就是“好”老板 TIP(Image Processing)最近的三任主编 UT Austin的Bovik Purdue的Jan P. Allebach,Charles Bouman; Bovik领域很宽,你看他很多方向都有很不错的成就,前段时间热衷QM,最近应该搞其他的多些?据说组里印度人蛮多; Allebach:图像,色彩什么的 Bouman:MAP重建,成像等等, 他们的研究方向最好看看主页,我说的不全,很容易搜到,他们都是Google的首条; 另外一些牛人: 哥大的Shih-fu Chang; 南加的QQ GUO UIUIC的Thomas S. Huang; 斯坦福的Bernd Girod takeo kanade (CMU)? Edward. Delp 牛人很多很多~列不完。。。这些算是学校好,导师也牛的吧; 此外,西北,Umich,RPI的图像视频组也不错 有些AP,可能不是特别有名,但是是将要出名的那种,要看命了吧。 中国的牛人: Changwen Chen, CSVT的主编,牛一个; Yao Wang, POLY, 牛一个; Ziqiang Xiong,TAMU,牛一个; Dapeng Wu, 北邮硕士毕业,UFL,牛~ 3, 领域介绍 (节选自IEEE,”The Golden Age of Imaging”) 写PS的时候可以参考:) The Golden Age of Imaging WE LIVE IN the golden age of imaging. Imaging research, development, and applications are growing at an astounding rate, and image-processing researchers can take credit for having created much of the enabling technologies that have fueled this growth. We are all familiar with the examples. The development of image and video coding standards, such as JPEG and MPEG, has enabled the web as a center for commerce and entertainment. Ubiquitous technologies, such as Direct TV, DVDs, BlueRay, and TiVo, depend on these standards; streaming internet video services, like iTunes’ recently announced movie rental feature, are well on their way to replacing traditional analog broadcast video. Other consumer products, such as home printers, digital cameras, and mobile video devices, have each been a major disruptive product enabled by fundamental innovation from image-processing researchers. In fact, image-processing research has not just impacted consumer products; it has changed the nature of scientific investigation and human healthcare. From the Mars Rover’s transmission of compressed digital video to the reconstruction of virus structure using cryo-electron microscopy, advanced digital imaging algorithms are at the heart of essential scientific investigation. In the field of healthcare, diagnostic imaging has revolutionized patient care. Any radiologist will tell you that volumetric CT and parallel acquisition MRI have changed medical imaging in the last decade by allowing dramatically faster and more accurate patient scans. These technologies have virtually eliminated the exploratory surgeries that were so common just a generation ago, and they promise to replace existing, invasive medical procedures, such as cardiac angiography. However, what your cardiologist may not know is that these technologies depended critically on the development of a variety of new volumetric image reconstruction algorithms that were developed by researchers in the imaging community. Image-processing research is clearly undergoing rapid growth. This growth is being fueled by a need for innovation in both established technologies and emerging applications. For example, Microsoft’s new HD compression format promises to have a great impact on the established world of digital photography and electronic imaging, and low-bit-rate video is becoming ever more important with the growth of mobile devices. At the same time, emerging research areas related to security, such as watermarking, surveillance, image forensics, and biometrics, are of rapidly growing importance. Moreover, entirely new fields are only starting to be realized. Research in image-based rendering and plenoptics is changing the way people think about conventional graphics and opens the possibility of imaging databases that could change our daily lives. If you doubt this possibility, you are welcome to drive past my home using Google’s Street View. Imaging researchers are uniquely equipped to solve this wide variety of problems because they understand the fundamental principals of applied math, algorithms, physical sensing, and human perception that lie at the core of these applications. As engineers, image-processing researchers also have the skills necessary to creatively assemble these pieces into solutions that address critical human needs. Clearly, imaging research has matured into a well-defined discipline with a core foundation of knowledge that cuts across application boundaries. It is a great field for students who enjoy both the difficult technical challenges and the opportunity to work collaboratively with multidisciplinary teams. The technical problems may be challenging, but experts in imaging have great opportunities, and their unique expertise can make them virtually irreplaceable in many organizations.
订阅后,新回复会通过你的通知中心匿名送达。
9 条回复
eeMars机器人#1 · 2009/6/2
另外,转载一个: 康涅狄格大学电子与计算机工程系教授招收博士生信息 他的主页:http://www.ee.uconn.edu/faculty.php?f_id=13 联系方式:Bahram.Javidi@UConn.edu 【 在 eeMars 的大作中提到: 】 : 为免以讹传讹,请大家多提提修改意见。 : 飞跃重洋 : 领域介绍 : ...................
eeMars机器人#2 · 2009/6/14
T.huang 2008年的paper: Gender recognition from body(2008) Liangliang Cao Mert Dikmen Yun Fu Thomas S. Huang Conference: ACM Multimedia Conference - MM SIFT-Bag kernel for video event analysis(2008) Xi Zhou Xiaodan Zhuang Shuicheng Yan Shih-fu Chang Mark Hasegawa-johnson Thomas S. Huang Conference: ACM Multimedia Conference - MM Image-Based Human Age Estimation by Manifold Learning and Locally Adjusted Robust Regression(2008) Guodong Guo Yun Fu Charles R. Dyer Thomas S. Huang Journal: IEEE Transactions on Image Processing Contextual motion field-based distance for video analysis(2008) Yadong Mu Shuicheng Yan Thomas S. Huang Bingfeng Zhou Journal: The Visual Computer - VC Latent Pose Estimator for Continuous Action Recognition(2008) Huazhong Ning Wei Xu Yihong Gong Thomas S. Huang Conference: European Conference on Computer Vision - ECCV Reconstruction and Recognition of Tensor-Based Objects With Concurrent Subspaces Analysis(2008) Dong Xu Shuicheng Yan Lei Zhang Stephen Lin Hong-jiang Zhang Thomas S. Huang Journal: IEEE Transactions on Circuits and Systems for Video Technology - TCSV Which "Apple" are you talking about ?(2008) Mandar Rahurkar Dan Roth Thomas S. Huang Conference: World Wide Web Conference Series - WWW Similarity Matching in Computer Vision and Multimedia(2008) Nicu Sebe Qi Tian Michael S. Lew Thomas S. Huang Journal: Computer Vision and Image Understanding - CVIU Real-Time Multimodal Human-Avatar Interaction(2008) Yun Fu Renxiang Li Thomas S. Huang Mike Danielsen Journal: IEEE Transactions on Circuits and Systems for Video Technology - TCSV Locally adaptive subspace and similarity metric learning for visual data clustering and retrieval(2008) Yun Fu Zhu Li Thomas S. Huang Aggelos K. Katsaggelos Journal: Computer Vision and Image Understanding - CVIU 我觉得从题目上和发的期刊上看,他最近的研究兴趣还是以video processing为主,加入视觉分析的一些原理,在交叉的点上做些东西; 另外,大家可以参考下大牛都喜欢投什么,就像俺帖子里说的: 会议的话,ACM的multimedia; 期刊:TIP,CSVT~ 大家加油~ 【 在 parodiushi 的大作中提到: 】 : T. Huang和T. Kanade,尤其是后者,更偏重Computer Vision,和Image/Video Processing还是有很大区别的,当然这些大牛做的东西非常广泛……
eeMars机器人#3 · 2009/6/14
这个我也觉得挺好的,不过没放在Tier 1里。。。回头补上吧。 这个好不好还是很主观的。。。我老板觉得在所有会议中ICASSP特别好,SPIE很好。。。 他的书架上ICASSP会议集若干,当然TIP也若干。。。 【 在 goldfish 的大作中提到: 】 : 没有cvpr?
eeMars机器人#4 · 2009/6/14
据说他在金融危机下逆势上扬,拿到一笔大funding,是真的么? 不过他那个MPEG 7太容易骗钱了。。。:) anyway,大牛人一个啊~ 【 在 traveller 的大作中提到: 】 : 哥大的Shih-fu Chang... : 这个名字不得不让我说点什么。这位台湾人现任哥大EE的chairman,平时交流感觉很和蔼,但要想进他的实验室可没那么容易--他还是比较倾向于Taiwaness。从大陆直接进他的实验室的两个人,一个是清华的牛人,另一个则在国内有n多的实习、竞赛经历,也是牛人。所以想作Shih-fu Chang的phd,先作牛人吧。所以如果被拒也没什么。
parodiushi机器人#5 · 2009/6/15
Video Processing 一般是指做编码的,主要是信号处理方向,多数属于ece或者ee,比较有名的教授比如Standford的Girod,USC的C.-C. Jay Kuo,Poly的Yao Wang(跟清华关系很好) 但是这个学科属于夕阳产业了 T. Huang算是Multimedia (MM是该领域最N的会),这些研究i需要很多IR、PR、CV的交叉,这些方向的老师多数在cs系,很多老师都是做CV的 另外paper不能光看题目,www是IR的著名会议 同样是Elsevier的期刊,T. Huang他们的CVIU是著名的CV刊物,如果是video processing应该投JVCI,当然最著名的是IEEE的TCSVT 【 在 eeMars 的大作中提到: 】 : T.huang 2008年的paper: : : Gender recognition from body(2008) : ...................
zixu1986机器人#6 · 2009/6/15
Multimedia的人喜欢做大系统 像mm就喜欢东西特别多的大系统 cv的人一般是针对一个特定问题 提出一种特定方法 multimedia一般是利用cv中的一些结果 不是直接做cv
eeMars机器人#7 · 2009/6/15
仁者见仁~ 不过编码只是video processing的一个分支~我认为~ 你说的这几个教授主要是编码和传输;这个方向其实有很多新的地方在做呢~不能算夕阳~但也不算朝阳吧; Kuo去年联系的时候说穷的都揭不开锅了;:) Yao Wang爆喜欢清华的,不是很喜欢其他学校的学生,Yao Wang所在的学校也不是很好,性价比不高; Girod不错的哦~人不错,钱也足~ Huang一直被叫做video processing的奠基人的,单说vision我觉得他不算最牛的。 另外,你看他的期刊有CSVT,还有TIP~很说明问题的~ 【 在 parodiushi 的大作中提到: 】 : Video Processing 一般是指做编码的,主要是信号处理方向,多数属于ece或者ee,比较有名的教授比如Standford的Girod,USC的C.-C. Jay Kuo,Poly的Yao Wang(跟清华关系很好) 但是这个学科属于夕阳产业了 : T. Huang算是Multimedia (MM是该领域最N的会),这些研究i需要很多IR、PR、CV的交叉,这些方向的老师多数在cs系,很多老师都是做CV的 : 另外paper不能光看题目,www是IR的著名会议 : ...................
lOnlyCaNcER机器人#8 · 2009/6/15
一个北理的,一个华科的,一个天大的,一个南大的,一个台湾新竹交大的 还有俩土耳其人 另外,Dapeng Wu主要还是比较偏通信更多吧~ 【 在 eeMars 的大作中提到: 】 : 她现在几个学生怎么样啊?
eeMars机器人#9 · 2009/6/16
不会这么惨淡吧。。。 Dapeng Wu年轻时候最有名的best paper是video传输的; 【 在 lOnlyCaNcER 的大作中提到: 】 : 一个北理的,一个华科的,一个天大的,一个南大的,一个台湾新竹交大的 : 还有俩土耳其人 : 另外,Dapeng Wu主要还是比较偏通信更多吧~