目标检测之CNN系列


排行榜

pascal VOC:http://host.robots.ox.ac.uk:8080/leaderboard/main_bootstrap.php

coco:http://mscoco.org/dataset/#detections-leaderboard

kitti:http://www.cvlibs.net/datasets/kitti/eval_object.php

Regionlets for Generic Object Detection

http://blog.csdn.net/maweifei/article/details/59078077

http://blog.csdn.net/maweifei/article/details/59078077


目标检测中的mAP是什么含义?

awesome-object-proposals

论文笔记之---Speed and accuracy trade-offs for modern convolutional object detectors

cs231n学习笔记-CNN-目标检测、定位、分割

基于深度学习的目标检测

ILSVRC2016目标检测任务回顾:图像目标检测(DET)

overfeat

深度学习(二十)基于Overfeat的图片分类、定位、检测-2014 ICLR

使用sklearn-theano来做object detection目标检测 (OverFeat)

深度学习研究理解:OverFeat:Integrated Recognition, Localization and Detection using Convolutional Networks




RCNN

读DL论文心得之RCNN

rcnn学习笔记:Rich feature hierarchies for accurate object detection and semantic segmentation 

R-CNN论文详解


DeepID-Net

DeepID-Net:multi-stage and deformable deep CNNs for object detection




selective search & Edge box

论文提要“Selective Search for Object Recognition” 

《Selective Search for object recognition》阅读笔记

Selective Search for Object Recognition(阅读)

What makes for effective detection proposals?

《Edge Boxes: Locating Object Proposals from Edges》读后感~

Oriented Object Proposals



其实·RPN也算一种

sppnet

SPPNet

论文笔记 《Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition》

读DL论文心得之SPP 

Spatial Pyramid Pooling in Deep Convolutional --- Spp_net

深度学习笔记(一)空间金字塔池化阅读笔记Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition

SPP pooling layer

深度学习(十九)基于空间金字塔池化的卷积神经网络物体检测-ECCV 2014

Spatial pyramid pooling (SPP)-net (空间金字塔池化)笔记



Fast RCNN

【目标检测】Fast RCNN算法详解 

读DL论文心得之Fast RCNN

论文提要"Fast R-CNN"

Fast R-CNN

论文笔记 《Fast R-CNN》

深度学习入门(二)Fast R-CNN

Fast-rcnn Notes 

Fast R-CNN笔记

论文笔记 Fast R-CNN细节

Fast R-CNN论文详解


Faster RCNN


【目标检测】Faster RCNN算法详解 

论文笔记:Faster R-CNN:Towards Real-Time Object Detection with Region Proposal Networks

Faster R-CNN论文笔记——FR

RCNN & SPP-net & Fast-RCNN & Faster-RCNN

物体检测-从RCNN到YOLO

Region Proposal Networks

Faster R-CNN学习笔记

Faster rcnn test浮点运算次数(卷积实现过程,Faster rcnn总体结构和参数)

bounding box regression

Faster R-CNN论文详解
feature map计算方法与faster-rcnn中roi映射到feature map的位置计算方法
Faster R-CNN

Faster RCNN-generate_anchors.py

faster-rcnn 之 RPN网络的结构解析

faster rcnn的特征图到原图区域映射问题

在CNN网络中roi从原图映射到feature map中的计算方法

针对Faster RCNN具体细节以及源码的解读之RoIPooling层

py-faster rcnn中rpn的3x3的滑框用卷积层来定义的是为什么?

Faster rcnn相关文章研究

Faster R-CNN论文详解

http://wenku.baidu.com/link?url=11zlcxN9p7d6ydhJHnDCBltkS9HEyLLZ0sBBgdwq7Oa02BsXeKRIBvDWcPVzKMFV8SOBtc9qdhTilohJ55MhUc7Ht7jDUiCh4yhn5xvRUYMiXX2T9lzG2zdYF5PDuFtn

Region Proposal Network

Single Bounding Box Regression

http://blog.csdn.net/qq_26898461/article/category/6204814

CTPN: Detecting Text in Natural Image with Connectionist Text Proposal Network

[目标检测] Faster R-CNN 深入理解 && 改进方法汇总



DeepBox

DeepBox: Learning Objectness with Convolutional Networks


YOLO

论文阅读笔记:You Only Look Once: Unified, Real-Time Object Detection

You Only Look Once: Unified, Real-Time Object Detection

论文阅读:You Only Look Once: Unified, Real-Time Object Detection

You Only Look Once: Unified, Real-Time Object Detection

http://blog.csdn.net/u012235274/article/category/6204386

幾個關於 ssd 與 yolo 無法理解的地方

图解YOLO

YOLO2

YOLO详解

YOLOv2 论文笔记

yolo v2 训练


DenseBox

DenseBox: Unifying Landmark Localization with End to End Object Detection


SSD

论文阅读:SSD: Single Shot MultiBox Detector

http://blog.csdn.net/u012235274/article/category/6366885

为什么SSD要比YOLO快?

ssd:single shot detector中的几个问题

SSD: Single Shot MultiBox Detector

基于深度学习的目标检测DET - SSD

SSD

DSSD: Deconvolutional Single Shot Detector 论文笔记



G-CNN

论文笔记 G-CNN: an Iterative Grid Based Object Detector



MultiPathNet

论文笔记 A MultiPath Network for Object Detection

https://github.com/facebookresearch/multipathnet


HyperNet

论文笔记 HyperNet: Towards Accurate Region Proposal Generation and Joint Object Detection

[论文阅读]HyperNet: Towards Accurate Region Proposal Generation and Joint Object Detection

HyperNet: Towards Accurate Region Proposal Generation and Joint Object Detection



LocNet

LocNet: Improving Localization Accuracy for Object Detection


OHEM

论文笔记 Bootstrapping Face Detection with Hard Negative Examples

Training Region-based Object Detectors with Online Hard Example Mining - cvpr 2016 oral

OHEM算法及Caffe代码详解

https://github.com/abhi2610/ohem

OHEM算法的Caffe实现



R-FCN

论文笔记 R-FCN: Object Detection via Region-based Fully Convolutional Networks

R-FCN:基于区域的全卷积网络来检测物体

R-FCN

[论文阅读]R-FCN: Object Detection via Region-based Fully Convolutional Networks

R-FCN: Object Detection via Region-based Fully Convolutional Networks

R-FCN: Object Detection via Region-based Fully Convolutional Networks

https://github.com/Orpine/py-R-FCN


MS-CNN


论文笔记 MSCNN:A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection

https://github.com/zhaoweicai/mscnn



PVANET

目标检测--PVANET: Deep but Lightweight Neural Networks for Real-time Object Detection

PVANET: Deep but Lightweight Neural Networks for Real-time Object Detection - arxiv 2016.08

论文笔记:PVANET: Deep but Lightweight Neural Networks for Real-time Object Detection



FPN

目标检测--Feature Pyramid Networks for Object Detection

FPN + DSSD 阅读笔记

FPN(feature pyramid networks)算法讲解

Feature Pyramid Networks for Object Detection 论文笔记



RRC

目标检测--Accurate Single Stage Detector Using Recurrent Rolling Convolution

Accurate Single Stage Detector Using Recurrent Ring Convolution



focal loss

focal loss



Object Detection方法汇总

R-CNN,SPP-NET, Fast-R-CNN,Faster-R-CNN, YOLO, SSD系列深度学习检测方法梳理

CVPR2016目标检测之识别效率篇:YOLO, G-CNN, Loc-Net

CVPR2016目标检测之识别精度篇:ResNet, ION, HyperNet,R-FCN

深度学习检测方法梳理

目标检测论文回顾

对话CVPR2016:目标检测新进展

基于深度学习的目标检测新作

【目标识别】深度学习进行目标识别的资源列表

【深度学习论文笔记】Deep Neural Networks for Object Detection

深度卷积神经网络在目标检测中的进展

目标检测“DPMs are CNNs”

视频目标检测 - Object Detection from Video Tubelets with Convolutional Neural Networks

目标检测“Exploit All the Layers: Fast and Accurate CNN Object Detector with Scale Dependent Pooling and”

Exploit All the Layers: Fast and Accurate CNN Object Detector with Scale Dependent Pooling 之再阅读

车辆检测”Learning to Detect Vehicles by Clustering Appearance Patterns“

Fast detection of multiple objects in traffic scenes with a common detection framework

3D Object Proposals for Accurate Object Class Detection

车辆检测“Integrating Context and Occlusion for Car Detection by Hierarchical And-Or Model”



论文提要“Filtered Channel Features for Pedestrian Detection”

论文提要“Taking a Deeper Look at Pedestrians”

论文提要“Pedestrian Detection aided by Deep Learning Semantic Tasks”



论文提要“Fast Feature Pyramids for Object Detection”


目标检测“Object Detection Using Generalization and Efficiency Balanced Co-occurrence Features”

BING

BING++: A Fast High Quality Object Proposal Generator at 100fps

综述:计算机视觉中RNN应用于目标识别

http://blog.csdn.net/zhangjunhit/article/category/6647655


Is Faster R-CNN Doing Well for Pedestrian Detection?

论文 Is Faster R-CNN Doing Well for Pedestrian Detection?探讨在行人检测领域Faster R-CNN是否有效,提出了RPN + Boosted Forest分类器。


基于R-CNN的多尺度改进方法概述


梳理基于R-CNN的多尺度改进方法,主要思路是提取多个层的feature进行卷积层的特征融合(即skip connections),涉及的方法有MultiPath Network,ION(Inside-Outside Net),HyperNet,PVANET及MS-CNN。

多尺度R-CNN论文笔记(1): A MultiPath Network for Object Detection

多尺度R-CNN论文笔记(2): Inside-Outside Net: Detecting Objects in Context with Skip Pooling and Recurrent Neural Networks

多尺度R-CNN论文笔记(3): HyperNet: Towards Accurate Region Proposal Generation and Joint Object Detection

多尺度R-CNN论文笔记(4): PVANET: Deep but Lightweight Neural Networks for Real-time Object Detection

多尺度R-CNN论文笔记(5): A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection

多尺度R-CNN论文笔记(6): Feature Pyramid Networks for Object Detection

Beyond Skip Connections: Top-Down Modulation for Object Detection阅读笔记



无proposal,基于回归的检测算法概述

开创性工作YOLO与后续改进,以及G-CNN,SSD等工作。
无proposal检测方法(1): You Only Look Once: Unified, Real-Time Object Detection

 无proposal检测方法(2): G-CNN: an Iterative Grid Based Object Detector

无proposal检测方法(3): SSD: Single Shot MultiBox Detector

YOLO9000: Better,Faster,Stronger(YOLO9000:更好,更快,更强)



和GAN结合

目标检测“A-Fast-RCNN: Hard Positive Generation via Adversary for Object Detection”

A-Fast-RCNN算法的Caffe实现

A-Fast-RCNN 论文笔记


弱监督/无监督

目标定位--Deep Self-Taught Learning for Weakly Supervised Object Localization

Is object localization for free? – Weakly Supervised Object Recognition with Convolutional Neural Networks

Weakly Supervised Deep Detection Networks

Weakly supervised localization of novel objects using appearance transfer

ContextLocNet: Context-Aware Deep Network Models for Weakly Supervised Localization

Weakly Supervised Cascaded Convolutional Networks

Weakly Supervised Object Localization with Progressive Domain Adaptation

https://github.com/jbhuang0604/WSL



PASCAL VOC数据集分析

VOC2007中包含9963张标注过的图片, 由train/val/test三部分组成, 共标注出24,640个物体。

VOC2007的test数据label已经公布, 之后的没有公布(只有图片,没有label)。
对于检测任务,VOC2012的trainval/test包含08-11年的所有对应图片。 trainval有11540张图片共27450个物体。
对于分割任务, VOC2012的trainval包含07-11年的所有对应图片, test只包含08-11。trainval有 2913张图片共6929个物体。

Pascal VOC 数据集介绍

VOC数据集mAP计算

Faster-R-CNN(Python).2:COCO数据集annotation内容

Bounding Box label code

Udacity Self-Driving 目标检测数据集简介与使用

图片标注工具LabelImg使用教程



spp_net实践

Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition
spp_net 代码实现过程
spp_net 在caffe上的实现
用caffe训练一个spp_net网络
spp_solver
spp_net 分类测试



RCNN实践

使用gpu(gtx1080) cudnn 5.1下编译faster rcnn

FAST RCNN安装配置精华

fast-rcnn配置运行demo.py(Ubuntu14.04)

faster-RCNN环境配置(Ubuntu14.04)

Faster RCNN 训练自己的检测模型

Py-faster-rcnn实现自己的数据train和demo

Faster-RCNN+ZF用自己的数据集训练模型(Python版本)

Faster-RCNN+VGG用自己的数据集训练模型

将数据集做成VOC2007格式用于Faster-RCNN训练

Faster R-CNN的安装、测试经历

Faster-rcnn 训练陨石坑检测

Fast RCNN训练自己的数据集 (2修改读写接口)

faster-rcnn安装,训练自己的数据

Faster R-CNN的安装、测试经历

浅析py-faster-rcnn中不同版本caffe的安装及其对应不同版本cudnn的解决方案

RCNN系列实验的PASCAL VOC数据集格式设置

ImageNet和PASCAL VOC图像描述的xml文件的解析、修改和生成

python生成VOC2007的xml代码

制作VOC类型数据集,生成txt,生成lmdb


Faster rcnn 安装、训练、测试、调试

搭建faster-rcnn进行目标检测的环境

faster rcnn:assert (boxes[:, 2] >= boxes[:, 0]).all()分析塈VOC2007 xml坐标定义理解

解决faster-rcnn中训练时assert(boxes[:,2]>=boxes[:,0]).all()的问题

GTX1080+Cuda8.0+Cudnnv5+caffe+faster-rcnn

py-faster-rcnn支持cuDNN V5的方法

faster rcnn +cudnn V5

faster rcnn +cudnn V5 不兼容问题的改进

faster rcnn demo.py:在一个窗口显示所有类别标注

TensorFlow 上基于 Faster RCNN 的目标检测

http://blog.csdn.net/10km/article/category/6816967

搭建faster-rcnn进行目标检测的环境

Faster-RCNN训练自己的数据集——备忘

Faster RCNN参数详解

faster rcnn multi GPU

https://www.google.com/search?sclient=psy-ab&site=&source=hp&btnG=Search&q=faster+rcnn+multi+gpu

https://github.com/rbgirshick/py-faster-rcnn/issues/143

https://github.com/bharatsingh430/py-R-FCN-multiGPU


Check rpn proposals in faster-rcnn

https://github.com/endernewton/tf-faster-rcnn

https://github.com/msracver/Deformable-ConvNets

https://github.com/bharatsingh430/Deformable-ConvNets/

yolo实践

yolo-darknet配置安装与测试

yolo-darknet实现自己数据的train和test

https://github.com/xingwangsfu/caffe-yolo

caffe 版本 yolo 过程记录

YOLO用自己的数据集训练模型

https://pjreddie.com/darknet/yolo/

http://guanghan.info/blog/en/my-works/train-yolo/

YOLOv2训练自己的数据集(VOC格式)

yolo训练之样本准备环节

使用YOLO训练自己的数据样本经验总结

yolo训练之训练结果评估环节

YOLO下一步:输出预测boundingbox做进一步处理

yolo模型的批量测试和位置输出

yolo的训练和测试




SSD实践

SSD的配置安装与测试

SSD train your own data

SSD: Signle Shot Detector 用于自然场景文字检测

SSD: Single Shot MultiBox Detector在Linux上的配置及运行

SSD(Single Shot MultiBox Detector):ubuntu16安装及训练自己的数据集(VOC2007格式)过程记录

SSD框架训练自己的数据集

以resnet作为前置网络的ssd目标提取检测

Single Shot Detection(SSD)bbox prediction相关的(超)参数

Anchors in SSD

SSD的图片预处理

SSD的Matching Strategy

https://github.com/balancap/SSD-Tensorflow

ssd.pytorch



pva net

caffe_pvanet,cuda7.5,VS2013

pva-faster-rcnn配置过程出现的问题(个人笔记)

自制作数据上的pva faster rcnn 训练




R-FCN实践

R-FCN配置(python版)

R-FCN+ResNet-50用自己的数据集训练模型(python版本)

R-FCN、SSD、YOLO2、faster-rcnn和labelImg实验笔记



你可能感兴趣的:(神经网络&深度学习,计算机视觉)