我是家家-CSDN博客

转载 python获取当前路径的方法

python获取当前路径的方法>>> import os>>> homedir = os.getcwd()>>> print homedirD:\python\testhttp://www.cnblogs.com/mfryf/p/3806135.html

2015-12-10 15:08:55 1651

原创 Caffe编译出错：/usr/bin/ld: cannot find lcblas

错误如下：BLAS或许是没有安装，或许是路径有问题。安装这个即可：

2015-11-17 11:14:56 8141 1

原创 Caffe编译出问题：This tool requires OpenCV; compile with USE_OPENCV.

解决方法：在Makefile.config文件中，添加：USE_OPENCV := 0即可正常编译。参考文件：https://groups.google.com/forum/m/#!topic/caffe-users/-zvZ9_RQapIhttps://github.com/BVLC/caffe/blob/master/Makefile.confi

2015-11-12 21:49:11 4685 7

原创修改paper的时候，一些标记符号

\newcommand{\cy}[1]{{\color{blue}#1}}\newcommand{\cyC}[1]{{\color{red}[#1: cy's comments to be removed!]}}

2015-09-09 20:04:48 1484

原创 latex 下划线

\uline{concat}

2015-09-09 20:02:32 7497

原创 latex 删除标记

\sout{spatisal}

2015-09-09 19:27:38 7809

转载 LaTex表格内单元格内容强制换行

/newcommand{/tabincell}[2]{/begin{tabular}{@{}#1@{}}#2/end{tabular}}%放在导言区%然后使用&/tabincell{c}{}&就可以在表格中自动换行%比如这么用/begin{tabular}{|c|c|}/hline1 & the first line ///hline2 & /tabincell

2015-09-08 16:38:09 9289

原创 Evernote 整理文档好工具可同步

https://evernote.com/下载地址。。。

2015-08-24 15:15:43 1070

翻译 arxiv about computer vision and pattern recognition

Address: http://arxiv.org/list/cs.CV/recentBe updated everyday, check it everyday.

2015-08-05 11:05:14 1565

翻译 Yann LeCun video: (What's Wrong with Deep Learning?)

video address: http://techtalks.tv/talks/whats-wrong-with-deep-learning/61639/

2015-08-05 11:03:10 1098

翻译 Image Super-Resolution Using Deep Convolutional Network （with demo）

Paper： Chao Dong, Chen Change Loy, Kaiming He, Xiaoou Tang, "Image Super-Resolution Using Deep Convolutional Networks", http://arxiv.org/abs/1501.0009 "For Creators", http://piapro.net/en_for_cr

2015-08-05 11:01:18 2227

翻译 RGB-D 图像（Semantic Pose using Deep Networks Trained on Synthetic RGB-D）

RGB-D图像其实是两幅图像：一个是普通的RGB三通道彩色图像，另一个是Depth图像。Depth图像类似于灰度图像，只是它的每个像素值是传感器距离物体的实际距离。通常RGB图像和Depth图像是配准的，因而像素点之间具有一对一的对应关系。利用Kinect可以采集RGB-D 图像。。。可以参考：Lai, K.; Bo, L.; Ren, X. & Fox, D. A large-sca

2015-08-05 10:56:43 2266

翻译 Recurrent network model for kinematic tracking

Model: Encoder-Recurrent-Decoder (ERD) model Task: for recognition and prediction of human body pose in videos and motion capture.... specially, motion capture generation, body pose labe

2015-08-04 12:25:27 974

原创 CUDA 打印输出 printf

示例： printf("%d\n",mask[ph * pooled_width + pw]);

2015-07-31 16:49:06 11706

转载解决办法：matlab Exception in thread "AWT-EventQueue-0" java.lang.OutOfMemoryError: Java heap

好几天没用matlab。。。竟然有问题了，不停的弹出。。。Exception in thread "AWT-EventQueue-0" java.lang.OutOfMemoryError: Java heap解决办法：在C:\Users\$username$\AppData\Roaming\MathWorks\MATLAB\R2013a目录下，找到matlab

2015-07-30 10:34:07 23889 5

翻译 Interaction between Visual cortex (biological) and computer vision

Deep hierarchies in the primate visual cortex: what can we learn for computer vision? TPAMI 2013 august.....

2015-07-24 11:15:28 769

转载 Semantic segmentation (with deconvolutional CNN) 作者公布源码和模型

作者主页：http://cvlab.postech.ac.kr/~hyeonwoonoh/看到这两篇工作，因为公布了好多有用的layer，融合了unpooling layer，batch normorlization layer....同时公布了protxt， model等文件。。1、learning deconvolutional network for semantic segme

2015-07-21 20:06:14 9140

转载 CIFAR数据库预处理 (python)

code说明和下载地址：http://deeplearning.net/software/pylearn2/tutorial/index.html#tutorial

2015-07-13 17:02:54 1528

转载 notepad++ 联想输入如何关掉

转自：http://www.zixue7.com/forum.php?mod=viewthread&tid=18932有时候这个提示好烦人，要去掉，怎么办？中文版：菜单栏，设置-----首选项---自动完成-----去掉勾英文版：

2015-06-22 10:21:17 14720 1

翻译 Image Super-Resolution using CNN

paper： Chao Dong, Chen Change Loy, Kaiming He, Xiaoou Tang. Learning a Deep Convolutional Network for Image Super-Resolution, in Proceedings of European Conference on Computer Vision (ECCV)

2015-06-20 17:22:34 7922

翻译 Zero-shot learning on the ImageNet of over 20000 classes

Devise: A deep visual-semantic embedding model (NIPS2013)Zero-Shot Learning by Convex Combination of Semantic Embeddings (http://arxiv.org/pdf/1312.5650.pdf)Mohammad Norouzi, Tomas Mikolov

2015-06-05 10:44:05 1721

翻译有公布code。Project-Out Cascaded Regression with an application to Face Alignment

Project-Out Cascaded Regression with an application to Face Alignment传统方法做的，效果还不错。

2015-06-03 15:41:16 2193 2

翻译 Some paper related to Image Deblurring

Generalized Video Deblurring for Dynamic ScenesKernel Fusion for Better Image DeblurringDeep Convolutional Neural Network for Image DeconvolutionLearning to Deblur (http://arxiv.

2015-06-03 15:22:11 1444

翻译 Scene Labeling with LSTM Recurrent Neural Networks

结果图：

2015-06-03 15:20:35 2239 1

转载 Deep Learning Tutorials

Deep Learning Tutorials:http://deeplearning.net/tutorial/

2015-06-03 15:08:40 921

翻译 STRIVING FOR SIMPLICITY: THE ALL CONVOLUTIONAL NET _CIFAR10 CIFAR100

网络结构图：结果还不错。。。

2015-06-03 14:27:33 2186

原创 Classify Images with Conceptor Network CIFAR-10 CIFAR-100 MNIST...

瞧

2015-06-03 11:25:48 1308

转载常用 blas 函数

转自：http://www.cnblogs.com/huashiyiqike/p/3886670.html

2015-06-02 18:42:57 946

原创 Large-scale Scene Understanding (LSUN) 竞赛

http://lsun.cs.princeton.edu/#organizers

2015-06-02 18:33:53 1857

转载 CVPR15 publiscation list

http://www.cv-foundation.org/openaccess/CVPR2015.py

2015-05-29 11:11:55 680

原创 CASIA WebFace Database 数据库

数据库地址：http://www.cbsr.ia.ac.cn/english/CASIA-WebFace-Database.html

2015-05-29 11:10:14 13423 2

翻译 SegNet: a deep convolutional encoder-decoder architecture for robust semantic pixel-wise labelling

框架：SegNet， a deep convolutional encoder-decoder architecture框架组成部分：A stack of encoders + a corresponding decoder stack + then feed into a softmax

2015-05-29 11:08:40 3066 1

原创 Object detection + CNN 两篇paper

1、object detection networks on convolutional feature maps2、Object detection via a multi-region & semantic segmentation-aware CNN model

2015-05-24 15:54:38 3587

转载如何编辑删除修改 pdf文件中的文字

要编辑修改PDF文件一般用三种软件，分别是：（1）Adobe Acrobat；（2）Foxit PDF Editor；（3）文电通。现介绍用Adobe Acrobat软件编辑PDF文件的方法：你先打开PDF文件，依次点击“工具”—“高级编辑工具”—“TouchUp文本工具（T）”，然后你用鼠标点击需要编辑的地方，就会出现一个长方形边框，在这个框内，你把光标移到需要编辑的位置，输入或删除文

2015-05-06 14:27:30 16488

转载 Li feifei How we're teaching computers to understand pictures

视频地址：http://www.ted.com/talks/fei_fei_li_how_we_re_teaching_computers_to_understand_pictures?language=enWhen a very young child looks at a picture, she can identify simple elements: "cat," "bo

2015-05-05 09:29:50 1258

原创 C/C++ 定义向量、赋值和使用

好多年前学的东西，都忘记了，只能重新开始了。头文件： #include 定义vector变量： vector mean_cy;vector变量的赋值： mean_cy.push_back(104); mean_cy.push_back(117); mean_cy.push_back(123);vector变量的使用： mean_cy[0]

2015-05-04 21:32:32 13149