<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[Maru_Lim的博客]]></title><description><![CDATA[machine learning/图像处理/目标识别检测跟踪]]></description><link>https://blog.youkuaiyun.com/Maru_Lim</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; Maru_Lim]]></copyright><item><title><![CDATA[用CenterNet对旋转目标进行检测]]></title><link>https://blog.youkuaiyun.com/Maru_Lim/article/details/110249781</link><guid>https://blog.youkuaiyun.com/Maru_Lim/article/details/110249781</guid><author>Maru_Lim</author><pubDate>Fri, 27 Nov 2020 23:24:29 +0800</pubDate><description><![CDATA[R-CenterNet——用CenterNet对旋转目标进行检测
代码
https://github.com/ZeroE04/R-CenterNet
demo


R-DLADCN(推荐)





R-ResDCN(主干网用的ResNet而不是DLA)





R-DLANet(未编译DCN的主干网)





DLADCN(原始CenterNet)





##前言


前段时间纯粹为了论文凑字数做的一个工作，本文不对CenterNet原理进行详细解读，如果你对CenterNet原理不了解，建议简单]]></description><category></category></item><item><title><![CDATA[算力3.5显卡服务器：子用户通过conda安装cuda9.0+cudnn7.6.4+torch1.2.0+配置远程jupyter+tensorflow[可选]]]></title><link>https://blog.youkuaiyun.com/Maru_Lim/article/details/109749052</link><guid>https://blog.youkuaiyun.com/Maru_Lim/article/details/109749052</guid><author>Maru_Lim</author><pubDate>Tue, 17 Nov 2020 18:46:40 +0800</pubDate><description><![CDATA[1. 清华源下载Anaconda3-2019.07-Linux-x86_64.sh，然后rz到服务器上，shAnaconda3-2019.07-Linux-x86_64.sh一路到头，有yes/no输入yes；

2. vi ~/.condarc


channels:
  - defaults
show_channel_urls: true
default_channels:
  - http://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main
  - h.]]></description><category></category></item><item><title><![CDATA[win7文件夹每次打开新的窗口/卸载ie后/同一窗口中打开每个文件也没用]]></title><link>https://blog.youkuaiyun.com/Maru_Lim/article/details/109374041</link><guid>https://blog.youkuaiyun.com/Maru_Lim/article/details/109374041</guid><author>Maru_Lim</author><pubDate>Thu, 29 Oct 2020 23:08:11 +0800</pubDate><description><![CDATA[首先如果你设置了““同一窗口中打开每个文件”也没用就是你卸载了IE浏览器，还跑去C盘把IE浏览器文件夹删了导致的。

以win7 64位为例。

在C:\Program Files\新建Internet Explorer文件夹，下载一个ieproxy.dll放进去，如果你是32位的就下载32位的，是64位的就下载64位的。

然后 win+r

regsvr32 "%SystemRoot%\System32\actxprxy.dll"

regsvr32 "%ProgramFiles%\Internet E]]></description><category></category></item><item><title><![CDATA[同时安装多版本VS以及配置不同版本的OpenCV]]></title><link>https://blog.youkuaiyun.com/Maru_Lim/article/details/103552153</link><guid>https://blog.youkuaiyun.com/Maru_Lim/article/details/103552153</guid><author>Maru_Lim</author><pubDate>Sun, 15 Dec 2019 19:49:46 +0800</pubDate><description><![CDATA[一、前言

其实多版本VS(visual studio)并不麻烦，只需要注意一定要先安装低版本的，再安装高版本的，高版本可以向下兼容，比如我自己是先安装了VS2010，后来又装了VS2015。

而配置OpenCV分两种情况，第一种就是你的两个版本的VS使用同一版本的OpenCV，那其实可以在装完VS2010后直接配置OpenCV，然后再安装VS2015，然后VS2015会默认和VS2010一样的...]]></description><category></category></item><item><title><![CDATA[2019算法面经汇总]]></title><link>https://blog.youkuaiyun.com/Maru_Lim/article/details/94574695</link><guid>https://blog.youkuaiyun.com/Maru_Lim/article/details/94574695</guid><author>Maru_Lim</author><pubDate>Wed, 03 Jul 2019 16:40:02 +0800</pubDate><description><![CDATA[作者：小小桔]]></description><category></category></item><item><title><![CDATA[剑指offer删除链表中重复结点]]></title><link>https://blog.youkuaiyun.com/Maru_Lim/article/details/92831865</link><guid>https://blog.youkuaiyun.com/Maru_Lim/article/details/92831865</guid><author>Maru_Lim</author><pubDate>Wed, 19 Jun 2019 15:10:52 +0800</pubDate><description><![CDATA[别人方法：
	

			class Solution:

			def deleteDuplication(self, pHead):

			# write code here

			ret= ListNode(0)

			real= ret

			pre= ret

			cur= pHe...]]></description><category></category></item><item><title><![CDATA[python2用python3的除法]]></title><link>https://blog.youkuaiyun.com/Maru_Lim/article/details/88681020</link><guid>https://blog.youkuaiyun.com/Maru_Lim/article/details/88681020</guid><author>Maru_Lim</author><pubDate>Wed, 20 Mar 2019 10:08:09 +0800</pubDate><description><![CDATA[python2默认/为整除，python3为浮点除。

若python2也想浮点除，则需要头文件添加：

from __feature__ import absolute_import

from __feature__ import division

from __feature__ import print_function
...]]></description><category></category></item></channel></rss>