- 博客(38)
- 资源 (8)
- 收藏
- 关注
原创 pytorch魔改data_set,帮助DataLoader实现enumerate(test_loader)载入image、target、name、oriimg
以项目pytorch-deeplab-xception为例:测试代码:https://github.com/jfzhang95/pytorch-deeplab-xception/issues/122 def test(self): self.model.eval() self.evaluator.reset() # tbar = ...
2019-10-31 20:27:20 5015
原创 项目pytorch-deeplab-xception为例,测试时怎么保存target、image:target.cpu().numpy()
一般性流程 ''' IPL转换为tensor _img = Image.open(os.path.join(self.img_dir, path)).convert('RGB') img = np.array(img).astype(np.float32).transpose((2, 0, 1)) img = t...
2019-10-31 11:09:06 2315
原创 点云配准算法-ICP改进 技术刘博客:http://www.liuxiao.org/ 泡泡机器人 一些PCL库学习博客或网站-几种主要的点云配准算法和开源软件库
课程:http://mesh.brown.edu/taubin/teaching.htmlqt and c++http://mesh.brown.edu/DGP/assignments.htmlhttp://mesh.brown.edu/byo3dhttp://mesh.brown.edu/DGP/coremesh/split.html图形学初学者,如何学习Polygon...
2019-10-30 11:29:57 2796
原创 一些机器学习框架-和logo
Tensorflowhttps://tensorflow.google.cn/腾讯:https://github.com/Angel-ML/angelhttps://github.com/microsoft/DMTKhttps://github.com/dmlc/tvmhttps://scikit-learn.org/stable/...
2019-10-30 10:56:48 307
原创 语义分割-实例分割论文下载地址
语义分割论文下载地址:2015 U-Net DeconvNet https://arxiv.org/pdf/1505.04366.pdf https://arxiv.org/pdf/1505.04597.pdf2016 SegNet FCN ENet DilatedNet PixelNet Refine...
2019-10-30 09:52:29 389
原创 PASCAL-VOC:指标问题 miou、pix-acc:以项目:pytorch-deeplab-xception为例和语义分割常用loss介绍:语义分割博客
博客:https://www.jianshu.com/u/ed6d4d31c5b0http://host.robots.ox.ac.uk/pascal/VOC/voc2012/http://host.robots.ox.ac.uk/pascal/VOC/深度学习图像分割(一)——PASCAL-VOC2012数据集(vocdevkit、Vocbenchmark_release)详...
2019-10-29 23:08:47 5176 2
原创 deeplab系列v1v2v3v3plus指标展示
deeplabv1:VOC2012-valhttps://arxiv.org/abs/1412.7062deeplabv2:VOC2012-valhttps://arxiv.org/abs/1606.00915deeplabv3:VOC2012-valhttps://arxiv.org/pdf/1706.05587v1.pdfdeeplabv3+:...
2019-10-29 18:29:09 329
原创 分割排行榜:COCO、cityscapes、VOC、MSRC-21
test提交格式:http://cocodataset.org/#format-resultshttp://cocodataset.org/#detection-evalhttps://competitions.codalab.org/https://competitions.codalab.org/competitions/5181http://cocodataset...
2019-10-29 17:19:39 2530
转载 项目awesome-semantic-segmentation-pytorch以及FCN、Unet、deeplabv1、deeplabv2、deeplabv3、deeplabv3+的网络
FCN、Unet、deeplabv1、deeplabv2、deeplabv3、deeplabv3+的网络https://blog.csdn.net/c_daofeng/article/details/90516691以DeepLabv3+架构为基础做图像分割(包含v1, v2, v3介绍)https://blog.csdn.net/weixin_42561002/article/det...
2019-10-28 13:29:21 4537 10
原创 查论文被引用
https://xs.glgoo.99lb.net/比如AffNet model implementation in PyTorch for ECCV2018 paper "Repeatability Is Not Enough: Learning Discriminative Affine Regions via Discriminability"Repeatability Is N...
2019-10-28 10:36:51 1083
原创 OpenMVS-OpenMVG-MVE-地址
OpenMVS: open Multi-View Stereo reconstruction libraryhttps://github.com/cdcseacave/openMVSOpenMVG (open Multiple View Geometry)https://github.com/openMVG/openMVGMulti-View Environmenthttps:...
2019-10-26 20:03:06 513
原创 ImageBasedModellingEduV1.0-master
ImageBasedModellingEduV1.0-masterhttps://github.com/swayfreeda/ImageBasedModellingEduV1.0改了bughttps://github.com/soarchorale/ImageBasedModellingEduV1.04个bug:https://github.com/swayfreeda/Ima...
2019-10-25 20:20:02 871 1
转载 PCL学习笔记-GitBook
https://mnewbie.gitbooks.io/pcl-notes/http://www.pclcn.org/
2019-10-25 16:23:18 205
转载 PCL中ply格式显示
参考:https://blog.csdn.net/weixin_34273481/article/details/86272869http://www.pcl-users.org/Reading-ply-wih-normal-data-td4025845.htmlPhotoneo和zivid两种传感器:Photoneoplyformat binary_little_endi...
2019-10-25 15:28:54 1091
原创 目标检测-分割论文中评价指标展示(AP、MIOU等)
检测Bounding Box Regression with Uncertainty for Accurate Object DetectionGeneralized Intersection over Union分割Dual Attention Network for Scene SegmentationReth...
2019-10-25 11:14:49 3642
原创 自动微分-深度学习系统-一个好博客
https://oldpan.me/以下是本系列文章的笔记来源:CSE 599W: Systems for ML AI-Sys Spring 2019http://dlsys.cs.washington.edu/schedulehttps://ucbrise.github.io/cs294-ai-sys-sp19/https://blog.csdn.net/tealex/article...
2019-10-25 10:46:11 490
原创 PCL阅读笔记
语法规则:http://www.pointclouds.org/documentation/advanced/pcl_style_guide.phphttp://www.pointclouds.org/documentation/advanced/pcl_style_guide.php#emacshttps://raw.githubusercontent.com/PointCloud...
2019-10-24 19:58:45 123
原创 PCL点云配准
初始尝试:https://www.jianshu.com/p/18d4270aad2ahttp://pointclouds.org/documentation/tutorials/interactive_icp.php#interactive-icp下载数据:https://github.com/PointCloudLibrary/datahttps://github.com...
2019-10-24 18:53:09 529
原创 Ubuntu-强制重启后,无法进入系统,显示 /dev/sdb6 *****files *****blocks
强制重启,进不去了,说多了都是泪,用了一下午的时间:df -hfdisk -l看看什么没有挂载上,我的是/home没有挂载上,home是/dev/sdb7,并不是/sdb6,他却显示/sdb6错误,管它呢,先修复fsck /dev/sdb6fsck /dev/sdb7我的会提示/dev/sdb6还在挂载中vim /etc/fstab那么 umount/dev...
2019-10-24 16:37:39 2263
原创 vcpkg安装pcl-Ubuntu16.04
安装X11sudo apt-get install libx11-dev libxext-dev libxtst-dev libxrender-dev libxmu-dev libxmuu-dev安装OpenGLsudo apt-get install build-essential libgl1-mesa-dev libglu1-mesa-dev键入如下命令来安装libglut-dev...
2019-10-23 18:16:01 3205
原创 PCL点云项目-配准-ubuntu16.04
sudo add-apt-repository ppa:v-launchpad-jochen-sprickerhof-de/pclsudo apt-get update会出错:cd /etc/apt/sources.list.dsudo mv v-launchpad-jochen-sprickerhof-de-ubuntu-pcl-xenial.list v-launchpad-...
2019-10-23 14:07:35 1945 1
原创 Python-offer
https://github.com/leeguandong/Interview-code-practice-python/tree/master/%E5%89%91%E6%8C%87offerhttps://github.com/Dengshunge/Target-Offer66# -*- coding:utf-8 -*-'''剪绳子题目描述给你一根长度为n的绳子,请...
2019-10-22 19:34:58 325
原创 pytorch-image-classification项目-修改最后的model.fc
直接可以修改def get_net(): #return MyModel(torchvision.models.resnet101(pretrained = True)) model = torchvision.models.resnet50(pretrained = True) #for param in model.parameters(): # ...
2019-10-21 16:10:35 1874 2
转载 数据转labelme,labelme转VOC,labelme转COCO
自己的转labelme# -*- coding:utf-8 -*-import argparseimport jsonimport osimport os.path as ospimport warningsimport copyimport numpy as npimport PIL.Imagefrom skimage import ioimport yamlim...
2019-10-21 14:49:48 2116
原创 记录-c语言面向对象
Glib GObjectGLIB、GTK中的“面向对象的继承”书籍:GTK+2.0编程范例GTK的官方网站上有编译好的Windows下Glib的开发包。http://ftp.gnome.org/pub/gnome/binaries/win32/glib/2.22/glib_2.22.2-1_win32.ziphttp://ftp.gnome.org/pub/gnome/binaries...
2019-10-20 12:05:39 200
转载 感受野计算
#https://blog.csdn.net/hzhj2007/article/details/79870936#https://link.zhihu.com/?target=https%3A//medium.com/mlreview/a-guide-to-receptive-field-arithmetic-for-convolutional-neural-networks-e0f51406...
2019-10-18 15:14:59 274
原创 mmdetection(v1.0rc0)-GTX2080遇到的问题-安装测试保存
./tools/dist_train.sh configs/faster_rcnn_r50_fpn_1x.py 3test.py configs/faster_rcnn_r50_fpn_1x.py checkpoints/epoch_1000.pth --out temp_pred/ouput.pkl --evaltrain.py configs/faster_rcnn_r5...
2019-10-16 19:41:40 5133 4
原创 pytorch-deeplab-xception项目问题-混淆矩阵和softmax特征类别设置,数据导入中心裁剪-保存分割图到TensorboardSummary
关于训练,测试时候,类别选择问题,首先是网络进入softmax之前的特征类别数目:#Define modelmodel = DeepLab(num_classes=self.nclass, backbone=args.backbone, output_stride=args.out_stride, ...
2019-10-15 19:58:12 1954 1
转载 vim问题汇总
vim黏贴代码格式混乱的解决方法https://blog.csdn.net/commshare/article/details/6215088
2019-10-15 10:17:33 170
转载 pytorch冻结BN和一些moudel
pytorch知识点--[ pytorch ] ——基本使用:(3) finetune冻结层操作 + 学习率超参数设Pytorch中的Batch Normalization layer踩坑https://blog.csdn.net/xiaojiajia007/article/details/90115174https://discuss.pytorch.org/t/free...
2019-10-15 09:47:57 4765
原创 事业单位招聘网址备份
http://81rc.81.cn/http://www.shiyebian.net/ningxia/http://www.nxpta.gov.cn/http://www.chinagwy.org/html/gdzk/ningxia/201911/89_326414.htmlhttp://nx.offcn.com/html/2019/10/75398.html...
2019-10-14 17:03:02 217
转载 pytorch中的num_workers和pin_memory
1.dataset:加载的数据集(Dataset对象)2.batch_size:batch size3.shuffle::是否将数据打乱4.sampler: 样本抽样,后续会详细介绍5.num_workers:使用多进程加载的进程数,0代表不使用多进程6.collate_fn: 如何将多个样本数据拼接成一个batch,一般使用默认的拼接方式即可7.pin_...
2019-10-14 16:10:44 4096
转载 远程访问服务器tensorboard-gelin
Python学习笔记4-本地浏览器登录远程服务器jupyterhttps://www.jianshu.com/p/02643a9cda85ssh -L 255.255.0.229 -X root@255.255.0.229 -p 30204https://www.jianshu.com/p/5865c9c39acc-X:开启X11转发功能-L listen-port:host:...
2019-10-13 13:21:06 285
原创 pytorch下载地址记录
http://mirrors.aliyun.com/pypi/simple/torch/http://mirrors.aliyun.com/pypi/simple/torchvision/https://pypi.doubanio.com/simple/torch/https://pypi.doubanio.com/simple/torchvision/https://...
2019-10-13 10:51:20 2626
转载 屏蔽广告
https://zhidao.baidu.com/question/988366216988866179.htmlhttps://www.cnblogs.com/aiyr/p/10500766.html
2019-10-13 10:03:31 155
原创 pytorch导出onnx格式时报错,RuntimeError: Failed to export an ONNX attribute, since it's not constant, please
这样,pytorch1.1.0就 可以转换onnx了https://blog.csdn.net/XDH19910113/article/details/100020205torch/onnx/symbolic_opset9.pyhttps://github.com/pytorch/pytorch/commit/11845cf2440c1abb0cb117bcee9532b...
2019-10-11 17:27:24 1289
转载 谷歌浏览器屏蔽百度热点搜索
https://www.cnblogs.com/aiyr/p/10500766.htmlhttps://zhidao.baidu.com/question/988366216988866179.html
2019-10-11 14:09:56 1228
GPU-ASIFT: A Fast Fully Affine-Invariant Feature Extraction Algorithm
2019-12-25
cmpmvs-v0.6.0-binary-win64-cuda.zip
2018-11-26
论文:关于三维重建和匹配
2018-10-19
LeNet-5 研习 2 附件
2018-10-07
空空如也
TA创建的收藏夹 TA关注的收藏夹
TA关注的人