边缘检测 (Edge-Detection) 论文、代码大汇总

边缘检测/边缘提取 论文、代码的大汇总。

Github上持续更新https://github.com/MarkMoHR/Awesome-Edge-Detection-Papers

examples

目录


1. 基于深度学习的方法

1.1 一般的边缘检测

方法简称论文发表处代码 / 项目链接
BDCNBi-Directional Cascade Network for Perceptual Edge DetectionCVPR 2019[code]
LPCBLearning to Predict Crisp BoundariesECCV 2018
AMH-NetLearning Deep Structured Multi-Scale Features using Attention-Gated CRFs for Contour PredictionNIPS 2017[code]
RCFRicher Convolutional Features for Edge DetectionCVPR 2017[code-caffe] [code-pytorch] [project]
CEDDeep Crisp BoundariesCVPR 2017[code]
COBConvolutional Oriented BoundariesECCV 2016[code] [project]
RDSLearning Relaxed Deep Supervision for Better Edge DetectionCVPR 2016
HFLHigh-for-Low and Low-for-High: Efficient Boundary Detection from Deep Object Features and its Applications to High-Level VisionICCV 2015
HEDHolistically-Nested Edge DetectionICCV 2015[code]
DeepEdgeDeepEdge: A Multi-Scale Bifurcated Deep Network for Top-Down Contour DetectionCVPR 2015
DeepContourDeepContour: A Deep Convolutional Feature Learned by Positive-sharing Loss for Contour DetectionCVPR 2015

1.2 物体轮廓提取

方法简称论文发表处代码 / 项目链接
CEDNObject Contour Detection with a Fully Convolutional Encoder-Decoder NetworkCVPR 2016[code-caffe] [code-TF]
Weakly Supervised Object BoundariesCVPR 2016

1.3 语义边缘检测 (包含分类)

方法简称论文发表处代码 / 项目链接
DFFDynamic Feature Fusion for Semantic Edge Detection1902.09104
SEALSimultaneous Edge Alignment and LearningECCV 2018[code]
CASENetCASENet: Deep Category-Aware Semantic Edge DetectionCVPR 2017[code]
datasetSemantic Contours from Inverse DetectorsICCV 2011[code]

1.4 遮挡边缘检测

方法简称论文发表处代码 / 项目链接
Occlusion Boundary Detection via Deep Exploration of ContextCVPR 2016

1.5 根据多帧进行边缘检测

方法简称论文发表处代码 / 项目链接
Boundary FlowBoundary Flow: A Siamese Network that Predicts Boundary Motion without Training on MotionCVPR 2018
LEGOLEGO: Learning Edge with Geometry all at Once by Watching VideosCVPR 2018[code]
Unsupervised Learning of EdgesCVPR 2016[code]

2. 传统方法

方法简称论文发表处代码 / 项目链接
SemiContourSemiContour: A Semi-supervised Learning Approach for Contour DetectionCVPR 2016
OEFOriented Edge Forests for Boundary DetectionCVPR 2015[code]
SEFast edge detection using structured forestsTPAMI 2015[code]
Edge BoxesEdge Boxes: Locating Object Proposals from EdgesECCV 2014[code]
PMICrisp Boundary Detection Using Pointwise Mutual InformationECCV 2014[code]
Sketch TokensSketch tokens: A learned mid-level representation for contour and object detectionCVPR 2013
SCGDiscriminatively Trained Sparse Code Gradients for Contour DetectionNIPS 2012
gPb-owt-ucmContour Detection and Hierarchical Image SegmentationTPAMI 2011
CannyA Computational Approach to Edge DetectionTPAMI 1986
#!/usr/bin/env python是一种在Python脚本中使用的特殊注释行,用于告诉操作系统在执行这个脚本时使用哪个Python解释器。这种写法的好处是可以避免用户没有将Python安装在默认路径(/usr/bin)下的情况。当系统看到这一行时,会先查找环境变量中的Python安装路径,然后调用对应路径下的解释器程序来执行脚本。相比之下,#!/usr/bin/python这种写法是直接指定了Python解释器的路径,相当于写死了路径。推荐使用#!/usr/bin/env python的写法,因为它可以根据环境设置自动寻找Python目录。\[1\]\[2\]\[3\] #### 引用[.reference_title] - *1* [Python脚本](https://blog.csdn.net/Dr_Guo/article/details/50635876)[target="_blank" data-report-click={"spm":"1018.2226.3001.9630","extra":{"utm_source":"vip_chatgpt_common_search_pc_result","utm_medium":"distribute.pc_search_result.none-task-cask-2~all~insert_cask~default-1-null.142^v91^insert_down1,239^v3^insert_chatgpt"}} ] [.reference_item] - *2* [#!/usr/bin/env python](https://blog.csdn.net/baidu_38869387/article/details/120078088)[target="_blank" data-report-click={"spm":"1018.2226.3001.9630","extra":{"utm_source":"vip_chatgpt_common_search_pc_result","utm_medium":"distribute.pc_search_result.none-task-cask-2~all~insert_cask~default-1-null.142^v91^insert_down1,239^v3^insert_chatgpt"}} ] [.reference_item] - *3* [Python脚本开头的#!/usr/bin/python](https://blog.csdn.net/qq_38380338/article/details/125637758)[target="_blank" data-report-click={"spm":"1018.2226.3001.9630","extra":{"utm_source":"vip_chatgpt_common_search_pc_result","utm_medium":"distribute.pc_search_result.none-task-cask-2~all~insert_cask~default-1-null.142^v91^insert_down1,239^v3^insert_chatgpt"}} ] [.reference_item] [ .reference_list ]
评论 5
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值