the principle of Calibration of cameras

some papers always describe they present a novel method of Intrinsic Calibration of cameras, but few say they can realize intrinsic and extrinsic calibration of cameras. How they implement that? And, most of people know the calibration is to obtain the accurate intrinsic. What’s the propose of extrinsic calibration of cameras?

Read More

Derivation of Kalman Filtering step by step

Basic knowledge

Bayes' theorem

Note that, $ P(x|y) $ is not only a conditional probability, but a Posterior probability for $x$, which means The probability of occurrence of $x$ after y occurs. $P(x)$ is not only a marginal probability, but a prior probability. $P(y|x)$ has three meanings, first is a conditional probability, and secondly, a Posterior probability for y, last but not least, is called by likelihood function. Also, $P(y)$ has three meanings, first is a marginal probability, and secondly, a prior probability, last but not least, is called by normalized constant.

Read More

『转』MIT教授关于学术写作的十大要诀

原作者:Jason Wayne 研之成理写作实验室(https://zhuanlan.zhihu.com/rationalscience-writing-lab)
原文链接:http://www.sztspi.com/archives/55211.html

前言

对于英语非母语的科研人员而言,学术写作的捷径是模仿,而在模仿之前必须了解学术写作的一些理论和要点。在此,小编打算写一个系列文章,选取一些功成名就的实操科学家,或者写作方面专家的文章、书中要点进行介绍。如果是文章,会附上原文链接;如果是书籍,会贴出封面。内容上,除了原文,会加上简单翻译和注解或一些书籍章节,以便拓宽、加深理解。今天主要分享麻省理工学院Ezra Zuckerman Sivan的一篇短文Tips to Article Writers。

原文链接(点击阅读原文可下载):
http://mitsloan.mit.edu/shared/ods/documents/?DocumentID=4448

Read More

警犬项目经验总结

警犬项目前前后后也已经做了1年出头了,虽然我对这种架构(所谓的单片机+树莓派这种机器人常见方案)用了很多年了,比较熟悉了,但是还是遇到了一个尚未探索到的地方,直到今天,我觉得大概做的是让自己达到了90%的满意度。

Read More

SVO详解

首先科普下计算机视觉和机器视觉

计算机视觉

主要是质的分析,比如分类识别,这是一个杯子那是一条狗。或者做身份确认,比如人脸识别,车牌识别。或者做行为分析,比如人员入侵等。

机器视觉

主要侧重对量的分析,比如用视觉去测量一个零件的直径。

Read More