2019年05月_seamanj

12月 11月 10月 09月 08月 07月 06月 05月 04月 03月 02月 01月

原创 Visualizing and Understanding Convolutional Networks

代码请见:https://github.com/amir-saniyan/ZFNet # Input: 224x224x3. with tf.name_scope('input'): self.X = tf.placeholder(dtype=tf.float32, shape=[N...

2019-05-29 01:30:32 227

原创 Mask R-CNN

code : https://github.com/facebookresearch/maskrcnn-benchmark?utm_source=mybridge&utm_medium=blog&utm_campaign=read_more

2019-05-31 03:15:34 143

原创 Fully Convolutional Networks for Semantic Segmentation

具体实现代码见:https://github.com/wkentaro/pytorch-fcn/blob/master/torchfcn/models/fcn16s.pyclass FCN16s(nn.Module): pretrained_model = \ osp.expanduser('~/data/models/pytorch/fcn16s_from_caff...

2019-05-31 03:04:59 469

原创 Feature pyramid networks for object detection

具体分析请见: https://medium.com/@jonathan_hui/understanding-feature-pyramid-networks-for-object-detection-fpn-45b227b9106cFPN is not an object detector by itself. It is a feature detector that works with ...

2019-05-31 02:55:45 316

原创 YOLOv3: An Incremental Improvement

具体代码见: https://github.com/ultralytics/yolov3def create_modules(module_defs): """ Constructs module list of layer blocks from module configuration in module_defs """ hyperparams = modu...

2019-05-30 19:17:48 875

原创 Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation

完整代码:https://github.com/jfzhang95/pytorch-deeplab-xceptionclass DeepLab(nn.Module): def __init__(self, backbone='resnet', output_stride=16, num_classes=21, sync_bn=True, freeze...

2019-05-30 18:50:53 411

原创 Dilated Residual Networks

完整代码请见: https://github.com/fyu/drnclass DRN(nn.Module): def __init__(self, block, layers, num_classes=1000, channels=(16, 32, 64, 128, 256, 512, 512, 512), ou...

2019-05-30 18:40:35 807

原创 CondenseNet: An Efficient DenseNet using Learned Group Convolutions

具体代码请见:https://github.com/ShichenLiu/CondenseNetclass CondenseNet(nn.Module): def __init__(self, args): super(CondenseNet, self).__init__() self.stages = args.stages se...

2019-05-29 20:55:10 424

原创 Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning

from __future__ import print_function, division, absolute_importimport torchimport torch.nn as nnimport torch.nn.functional as Fimport torch.utils.model_zoo as model_zooimport osimport sys__a...

2019-05-29 20:41:06 598

原创 Rethinking the Inception Architecture for Computer Vision

/home/seamanj/Software/anaconda3/bin/python3.7 /home/seamanj/Workspace/AlexNet/main.pyDownloading: "https://download.pytorch.org/models/inception_v3_google-1a9a5a14.pth" to /home/seamanj/.cache/t...

2019-05-29 20:34:11 819

原创 Xception: Deep Learning with Depthwise Separable Convolutions

代码请见:https://github.com/Cadene/pretrained-models.pytorch/blob/master/pretrainedmodels/models/xception.py"""Ported to pytorch thanks to [tstandley](https://github.com/tstandley/Xception-PyTorch)@a...

2019-05-29 20:20:39 682

原创 SQUEEZENET: ALEXNET-LEVEL ACCURACY WITH 50X FEWER PARAMETERS AND 0.5MB MODEL SIZE

/home/seamanj/Software/anaconda3/bin/python3.7 /home/seamanj/Workspace/AlexNet/main.pyDownloading: "https://download.pytorch.org/models/squeezenet1_0-a815701f.pth" to /home/seamanj/.cache/torch/ch...

2019-05-29 20:12:50 288

原创 Squeeze-and-Excitation Networks

'''SENet in PyTorch.SENet is the winner of ImageNet-2017. The paper is not released yet.'''import torchimport torch.nn as nnimport torch.nn.functional as Fclass BasicBlock(nn.Module): de...

2019-05-29 20:07:43 269

原创 Aggregated Residual Transformations for Deep Neural Networks

/home/seamanj/Software/anaconda3/bin/python3.7 /home/seamanj/Workspace/AlexNet/main.pyDownloading: "https://download.pytorch.org/models/resnext50_32x4d-7cdf4587.pth" to /home/seamanj/.cache/torch/ch...

2019-05-29 20:00:17 249

原创 ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design

class ShuffleNetV2(nn.Module): def __init__(self, net_size): super(ShuffleNetV2, self).__init__() out_channels = configs[net_size]['out_channels'] num_blocks = configs[ne...

2019-05-29 19:53:03 704

原创 CNN 总结模型归类

部分模型实现代码:https://github.com/kuangliu/pytorch-cifar/tree/master/models

2019-05-29 19:42:43 202

原创 ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices

具体见: https://github.com/kuangliu/pytorch-cifar/tree/master/modelsclass ShuffleNet(nn.Module): def __init__(self, cfg): super(ShuffleNet, self).__init__() out_planes = cfg['out_p...

2019-05-29 19:34:54 213

原创 MobileNetV2: Inverted Residuals and Linear Bottlenecks

推荐阅读:https://blog.csdn.net/kangdi7547/article/details/8143157216->96->24, 为什么中间有96怎么确定, 由t来确定, 这里t为6, 所以16*6 =96MobileNetV2( (features): Sequential( (0): ConvBNReLU( (0): Conv2d(3...

2019-05-29 19:09:17 560

原创 MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications

具体实现代码见:https://github.com/marvis/pytorch-mobilenetclass Net(nn.Module): def __init__(self): super(Net, self).__init__() def conv_bn(inp, oup, stride): return nn.Seq...

2019-05-29 18:42:48 163

原创 Convolutional Neural Networks with Alternately Updated Clique

具体代码见:https://github.com/iboing/CliqueNetclass build_cliquenet(nn.Module): def __init__(self, input_channels, list_channels, list_layer_num, if_att): super(build_cliquenet, self).__init_...

2019-05-29 18:32:15 203

原创 Densely Connected Convolutional Networks

DenseNet( (features): Sequential( (conv0): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False) (norm0): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track...

2019-05-29 18:01:12 204

原创 Deep Residual Learning for Image Recognition

resnet也有各种变种ResNet( (conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False) (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)...

2019-05-29 03:47:42 139

原创 VERY DEEP CONVOLUTIONAL NETWORKS FOR LARGE-SCALE IMAGE RECOGNITION

VGG( (features): Sequential( (0): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) (1): ReLU(inplace) (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1,...

2019-05-29 03:27:34 140

原创 Going deeper with convolutions

import torchvision.models as modelsif __name__ == '__main__': googlenet = models.googlenet(pretrained=True) print(googlenet)GoogLeNet( (conv1): BasicConv2d( (conv): Conv2d(3, 64, k...

2019-05-29 03:00:04 131

原创 torchvision datasets and models

AlexNetVGGResNetSqueezeNetDenseNetInception v3GoogLeNetShuffleNet v2MobileNet v2ResNeXt

2019-05-29 02:15:34 221 1

原创 Learning to Refine Object Segments

源代码里面的版本看下源代码class SharpMask(nn.Module): def __init__(self, config=default_config, context=True): super(SharpMask, self).__init__() self.context = context # with context ...

2019-05-28 23:31:30 583

原创 Learning to Segment Object Candidates

看下源代码class DeepMask(nn.Module): def __init__(self, config=default_config, context=True): super(DeepMask, self).__init__() self.config = config self.context = context # w...

2019-05-28 20:25:47 244

原创 imagenet classification with deep convolutional neural networks

3.1这里用了ReLU, 而不是sigmoid或者是tanh, 在文章中称sigmoid或tanh为saturating nonlinearities. 为啥呢, 具体可见https://www.quora.com/Why-would-a-saturated-neuron-be-a-problem这里我挑重点The logistic has outputs between 0 and 1, ...

2019-05-28 20:11:21 208

原创 Gradient-Based Learning Applied to Document Recognition

对应codeimport torch.nn as nnfrom collections import OrderedDictclass LeNet5(nn.Module): """ Input - 1x32x32 C1 - 6@28x28 (5x5 kernel) tanh S2 - 6@14x14 (2x2 kernel, stride 2) ...

2019-05-28 17:12:47 1687

转载解决Ubuntu下Qt Creator无法输入中文

环境：ubuntu16.04fcitx的输入法Qt Creator 5.10.1首先需要安装sudo apt-get install fcitx-frontend-qt5然后将：/usr/lib/x86_64-linux-gnu/qt5/plugins/platforminputcontexts/libfcitxplatforminputcontextplugin.so拷贝到：~...

2019-05-21 17:18:32 310

原创 qtcreator 用cmake找不到userPath的解决方法

在写CUDA程序的时候,即使在cmake用了include_directories, 然后ctrl+shift+F12一查看还没有UserPath, 尽管可以编译成功, 但是错误的提示找不到头文件让程序写起来十分头痛. 总结一下原因, 有可能是qtCreator的bug, 但是我看了一下,大概是因为只有.h 和 .cu文件没有.cpp文件造成的, 解决方法就是加一个空的同名的.cpp就可以了...

2019-05-17 04:17:12 823

原创 check eigen version

Eigen/src/Core/util/Macros.h

2019-05-13 20:44:09 1023

原创 install anttweakbar on ubuntu

cd AntTweakBar/srcmakesudo cp ../include/* /usr/local/includesudo cp ../lib/libAntTweakBar.* /usr/local/lib

2019-05-11 15:33:46 433

原创 build tut framework on ubuntu

download http://mrzechonek.github.io/tut-framework/python2 waf configuresudo python2 waf installdone!using in cmakefind_package(PkgConfig REQUIRED)pkg_search_module(tut REQUIRED tut)

2019-05-07 02:33:51 102

原创 ubuntu下pkgconfig的搜索路径

/usr/lib/pkgconfig/usr/local/lib/pkgconfig

2019-05-07 02:25:48 1227

原创 cmake设置了CMAKE_MODULE_PATH,include找不到文件

设置了CMAKE_MODULE_PATH后, 如何要包含某一文件include(CudaComputeTargetFlags) #注意这里不需要加.cmake后缀, 否则会找不到文件https://stackoverflow.com/questions/3249459/for-the-cmake-include-command-what-is-the-difference-between-a-...

2019-05-03 04:06:12 5790

原创 cmake and cmake-gui读的环境变量不一致

在配OpenNI2的时候, 导出变量到~/.bashrc后, 发现cmake和cmake-gui读出的变量不一致,仔细一看cmake读的变量是从~/.bashrc文件里面的读的cmake-gui是从~/.profile文件里面读的可以在cmake文件里面加入execute_process( COMMAND ${CMAKE_COMMAND} -E environment)看细节...

2019-05-02 21:37:46 1008

原创 Efficient Inference in Fully Connected CRFs with Gaussian Edge Potential

这里我们把每列当成像素, 每行当成不同的label, 这里有四种label.然后我们需要算在每个点比如第一列第二行的点则为Q1(x1=第二种label)Q_1(x_1 = 第二种label)Q1(x1=第二种label), 通过算法一达到收敛确定出每个点的Q值, 最后对每列取最大值所在行,则为该像素的label这里主要需要确定unary potential and pairwise po...

2019-05-01 05:27:22 679

原创 Co-Fusion: Real-time Segmentation, Tracking and Fusion of Multiple Objects

这篇文章其实就是几篇文章凑在了一起.fusion的部分用的是ElasticFusion: Dense SLAM without a pose graph这篇文章实现了local loop closure and global loop closure然后他的segmentation有两种, 一种是MOTION SEGMENTATION, 它是基于超像素来做这种segmentation. 对...

2019-05-01 04:51:54 1263