存档

作者存档

必须激动一把:学术论文里看到魅族M8

2010年11月24日 2 条评论

外国煤油很给力啊!这篇文章:

Wide Area Localization on Mobile Phones

Authors Arth Clemens, Wagner Daniel, Klopschitz Manfred, Irschara Arnold , Schmalstieg Dieter
Appeared in Proceedings of Int. Symposium on Mixed and 2009 (ISMAR´09)

文章这里下载。摘要:

We present a fast and memory efficient method for localizing a mobile user’s 6DOF pose from a single camera image. Our approach registers a view with respect to a sparse 3D point reconstruction. The 3D point dataset is partitioned into pieces based on visibility constraints and occlusion culling, making it scalable and efficient to handle. Starting with a coarse guess, our system only considers features that can be seen from the user’s position. Our method is resource efficient, usually requiring only a few megabytes of memory, thereby making it feasible to run on low-end devices such as mobile phones. At the same time it is fast enough to give instant results on this device class. m8 in

同样是煤油,人家可以拿M8来做实验写paper,还HLL地在paper里给M8两个特写,差距啊!

就是不知道这帮哥们怎么买到M8的,而且还是白色后盖的。

不知道囧王看到这个啥表情。

更坚定了我买M9的信念,嘿嘿!

分类: 新闻 标签:

文档图像对齐以及AR

2010年11月24日 6 条评论

去年曾经很需要的一个算法,今天无意中遇到了。

图像配准早就是一个很easy的技术了,但想想近似于二值图像的文档图像如何配准?

直接上SIFT,SURF这些经典算法?No!曾经有人试验过,结果是有效的feature point非常非常之少,近似于黑白的图像和相似的结构,那些直方图特征搞不定啊。

feature points

(a)和(b)分别是SURF和文章的算法

registration

出自同一个很NB的作者的这两篇文章提供了比较好的解决办法。具体解决办法看论文(1 2)和video(1 2)吧。很nice作者还提供了源码

  1. Hideaki Uchiyama, Julien Pilet and Hideo Saito, “On-line document registering and retrieving system for annotation overlay,” 1st ACM Augmented Human International Conference (AH), No.23, Apr. 2010. (oral)
  2. Hideaki Uchiyama and Hideo Saito, “Augmenting text document by on-line learning of local arrangement of keypoints,” 8th IEEE and ACM International Symposium on Mixed and (ISMAR), pp.95-98, Oct. 2009. (oral)

CCAV也AR

2010年6月19日 13 条评论

正在看央视五套的世界杯节目“豪门盛宴”,惊奇地发现里面用到了增强现实,用来在舞台中间显示一些虚拟的东西,比如谈到英格兰队时在舞台中间虚拟地显示一些新闻,赛前在舞台中间显示本场比赛的体育场的模型。虽然手上没有相机,好在Google到一些。有人说mark是场地中央的那个团,对此表示赞同。此外里面还有一些3d技术来分析球员站位,不知道是世界杯官方提供的还是央视自己做的,很炫很实用。

看下图,主持人跑到体育场后面了。

阅读全文…

无意中Google到的

2010年6月6日 4 条评论

http://www.statinternet.com/domain/cvchina.info


.info evaluated about 5120$ USD

We estimate the value of cvchina.info according to the number of visitors during the history. It is located in , ,and IP Address is . It has about 899 unique users every day.

Recent Data of cvchina.info

Users Information
Domain Name: cvchina.info Domain Name Worth: $5120 USD
Daily Visitors: 899 Daily Page Views: 5770
Alexa Rank: 813866 Users from Search Engine: 112

Main Page Information

Main Page Information
Title: 增强视觉 |
Keywords: 增强现实, , 公司介绍, siggraph, goggles, where 2.0, Image Edit, 机器学习, 图像编辑, ML, 虚拟化妆镜, face detection, 随机相机, CG, 虚拟换衣镜, Photo Generate, captcha, 图像标注, techxav, Fjorg Competition, alpha matting, 反人脸检测, SmartBoxes, machine learning, oMoby, 人脸检测, lomo, SIGGRAPH Asia, crowdsourcing, four square, New Orleans, Sketch2Photo, 逆向工程, IQ Engines, geo-location, Image Pattern, 云计算, NYU, 众包, 计算机视觉 增强现实 augmented reality
Description: This website doesn’t have description

Whois Information

Whois Information
Whois data for: cvchina.info
Access to INFO WHOIS information is provided to assist persons in
determining the contents of a domain name registration record in the
Afilias registry database. The data in this record is provided by
Afilias Limited for informational purposes only, and Afilias does not
guarantee its accuracy. This service is intended only for query-based
access. You agree that you will use this data only for lawful purposes
and that, under no circumstances will you use this data to: (a) allow,
ena

微软年度技术展

2010年3月3日 没有评论

激光打蚊子

2010年3月2日 2 条评论

应该不是基于视觉,但很好玩。
夏夜,卧室内,仰望天花板,一起来看蚊子雨,还有昆虫版的飞毛腿与爱国者的空中大战。
被蚊子咬了的可以看下面的视频泄愤。

该“武器”每秒可击毙50到100只蚊子。除了速度快之外,该激光器还很精准,它能区别蝴蝶和蚊子,也能分辨雌蚊子和雄蚊子。只有雌蚊子(翅膀低频拍打)才会叮咬人类。
“一夜间你能杀死几十亿只蚊子,还不会伤害蝴蝶。”灭蚊激光器的最终目标是使用激光为宅所和医院周围创造保护性屏障,此外,还可用于农场,替代杀虫剂。

原文:
http://online.wsj.com/article/SB123680870885500701.html
http://www.wired.com/epicenter/2010/02/death-star-laser-zaps-mosqitoes-dead/

蚊子被击中的慢放视频:
http://intellectualventureslab.com/wp-content/videos/Mosquito_Shootdown_Sequence

分类: 新闻 标签: