【论文汇总】2D目标检测文章汇总，持续更新_target-aware dual adversarial learning-程序员宅基地

记录自己比较感兴趣的2D目标检测文章

1.模型架构相关

Date	Pub.	Title	Code
2022	ECCV	ObjectBox: From Centers to Boxes for Anchor-Free Object Detection	https://github.com/mohsenzand/objectbox
2021	MM	Disentangle Your Dense Object Detector	https://github.com/zehuichen123/DDOD
2021	CVPR	VarifocalNet: An IoU-aware Dense Object Detector	https://github.com/hyz-xmaster/VarifocalNet
2021	CVPR	Sparse R-CNN: End-to-End Object Detection with Learnable Proposals	https://github.com/PeizeSun/SparseR-CNN
2020	ECCV	Dynamic R-CNN: Towards High Quality Object Detection via Dynamic Training	https://github.com/hkzhang95/DynamicRCNN
2020	ECCV	Side-Aware Boundary Localization for More Precise Object Detection	https://github.com/open-mmlab/mmdetection
2020	CVPR	CentripetalNet: Pursuing High-quality Keypoint Pairs for Object Detection	https://github.com/KiveeDong/CentripetalNet
2020	CVPR	NAS-FCOS: Fast Neural Architecture Search for Object Detection	https://github.com/open-mmlab/mmdetection
2020	TIP	FoveaBox: Beyond Anchor-based Object Detector	https://github.com/taokong/FoveaBox
2020	NIPS	RepPoints V2: Verification Meets Regression for Object Detection	https://github.com/Scalsol/RepPointsV2
2019	ICCV	RepPoints: Point Set Representation for Object Detection	https://github.com/microsoft/RepPoints
2019	NIPS	Cascade RPN: Delving into High-Quality Region Proposal Network with Adaptive Convolution	https://github.com/thangvubk/Cascade-RPN
2019	ICCV	FCOS: Fully Convolutional One-Stage Object Detection	https://github.com/tianzhi0549/FCOS
2019	ICCV	Scale-Aware Trident Networks for Object Detection	https://github.com/open-mmlab/mmdetection
2019	CVPR	Libra R-CNN: Towards Balanced Learning for Object Detection	https://github.com/open-mmlab/mmdetection
2019	CVPR	Objects as Points	https://github.com/xingyizhou/CenterNet
2019	CVPR	Region Proposal by Guided Anchoring	https://github.com/open-mmlab/mmdetection
2019	CVPR	Grid R-CNN	https://github.com/open-mmlab/mmdetection
2018	ECCV	CornerNet: Detecting Objects as Paired Keypoints	https://github.com/princeton-vl/CornerNet
2018	CVPR	Cascade R-CNN: High Quality Object Detection and Instance Segmentation	https://github.com/zhaoweicai/cascade-rcnn
2017	ICCV	Focal Loss for Dense Object Detection	https://github.com/open-mmlab/mmdetection
2016	ECCV	SSD: Single Shot MultiBox Detector	https://github.com/weiliu89/caffe
2015	NIPS	Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks	https://github.com/ShaoqingRen/faster_rcnn
2015	ICCV	Fast R-CNN	https://github.com/rbgirshick/fast-rcnn

2.YOLO系列

Date	Pub.	Title	Code
2023	—	YOLOV8	https://github.com/ultralytics/ultralytics
2022	—	DAMO-YOLO	https://github.com/tinyvision/DAMO-YOLO
2022	arXiv	YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors	https://github.com/wongkinyiu/yolov7
2022	—	Yolov6	https://github.com/meituan/YOLOv6
2022	arXiv	PP-YOLOE: An evolved version of YOLO	https://github.com/PaddlePaddle/PaddleDetection
2021	Tech Report	YOLOX: Exceeding YOLO Series in 2021	https://github.com/Megvii-BaseDetection/YOLOX
2021	aiXiv	You Only Learn One Representation: Unified Network for Multiple Tasks	https://github.com/WongKinYiu/yolor
2021	CVPR	Scaled-YOLOv4: Scaling Cross Stage Partial Network	https://github.com/WongKinYiu/ScaledYOLOv4
2021	arXiv	PP-YOLOv2: A Practical Object Detector	https://github.com/PaddlePaddle/PaddleDetection
2020	arXiv	PP-YOLO: An Effective and Efficient Implementation of Object Detector	https://github.com/PaddlePaddle/PaddleDetection
2020	—	Yolov5	https://github.com/ultralytics/yolov5
2020	arXiv	Yolov4: Optimal speed and accuracy of object detection	https://github.com/AlexeyAB/darknet
2018	Tech Report	YOLOv3: An Incremental Improvement	—
2017	CVPR	YOLO9000: Better, Faster, Stronger	—
2016	CVPR	You Only Look Once: Unified, Real-Time Object Detection	—

3.分类与回归不一致问题

Date	Pub.	Title	Code
2021	ICCV	TOOD: Task-aligned One-stage Object Detection	https://github.com/fcjian/TOOD
2021	ICCV	Reconcile Prediction Consistency for Balanced Object Detection	—
2021	ICCV	Mutual Supervision for Dense Object Detection
2020	CVPR	Rethinking Classification and Localization for Object Detection	https://github.com/wuyuebupt/doubleheadsrcnn
2020	CVPR	Multiple Anchor Learning for Visual Object Detection	https://github.com/KevinKecc/MAL

4.标签分配

Date	Pub.	Title	Code	备注
2023	CVPR	One-to-Few Label Assignment for End-to-End Dense Detection	https://github.com/strongwolf/o2f
2022	CVPR	A Dual Weighting Label Assignment Scheme for Object Detection	https://github.com/strongwolf/dw
2021	Neurocomputing	LLA: Loss-aware Label Assignment for Dense Pedestrian Detection	https://github.com/Megvii-BaseDetection/LLA	笔记
2021	CVPR	IQDet: Instance-wise Quality Distribution Sampling for Object Detection	—
2021	CVPR	OTA: Optimal Transport Assignment for Object Detection	https://github.com/Megvii-BaseDetection/OTA
2020	ECCV	Probabilistic Anchor Assignment with IoU Prediction for Object Detection	https://github.com/kkhoot/PAA
2020	CVPR	Bridging the Gap Between Anchor-based and Anchor-free Detection via Adaptive Training Sample Selection	https://github.com/sfzhang15/ATSS
2020	arXiv	AutoAssign: Differentiable Label Assignment for Dense Object Detection	https://github.com/Megvii-BaseDetection/AutoAssign
2019	NIPS	FreeAnchor: Learning to Match Anchors for Visual Object Detection	https://github.com/zhangxiaosong18/FreeAnchor
2019	CVPR	Region Proposal by Guided Anchoring	—
2018	NIPS	MetaAnchor: Learning to Detect Objects with Customized Anchors	-

5.DETR系列

Date	Pub.	Title	Code
2023	arXiv	RT-DETR: DETRs Beat Yolos on Real-time Object Detection	https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/rtdetr#%E6%A8%A1%E5%9E%8B
2022	CVPR	DN-DETR: Accelerate DETR Training by Introducing Query DeNoising	https://github.com/fengli-ust/dn-detr
2021	ICLR	Deformable DETR: Deformable Transformers for End-to-End Object Detection	https://github.com/fundamentalvision/Deformable-DETR
2021	ICCV	Rethinking Transformer-based Set Prediction for Object Detection	https://github.com/edward-sun/tsp-detection
2021	ICCV	Dynamic DETR: End-to-End Object Detection with Dynamic Attention	—
2020	ECCV	End-to-End Object Detection with Transformers	https://github.com/facebookresearch/detr

6.知识蒸馏

Date	Pub.	Title	Code
2022	CVPR	Localization Distillation for Dense Object Detection	https://github.com/HikariTJU/LD
2022	CVPR	Focal and Global Knowledge Distillation for Detectors	https://github.com/yzd-v/FGD
2022	WACV	Improving Object Detection by Label Assignment Distillation	https://github.com/cybercore-co-ltd/CoLAD
2021	CVPR	Distilling Object Detectors via Decoupled Features	https://github.com/ggjy/DeFeat.pytorch
2021	CVPR	General Instance Distillation for Object Detection	—

7.FPN相关

Date	Pub.	Title	Code
2022	ECCV	You Should Look at All Objects	https://github.com/charlespikachu/yslao
2021	CVPR	You Only Look One-level Feature	https://github.com/megvii-model/YOLOF
2019	CVPR	Feature Selective Anchor-Free Module for Single-Shot Object Detection	https://github.com/open-mmlab/mmdetection

8.小目标检测系列

Date	Pub.	Title	Code
2023	arXiv	TinyDet: Accurate Small Object Detection in Lightweight Generic Detectors	https://github.com/hustvl/TinyDet
2022	ECCV	RFLA: Gaussian Receptive Field based Label Assignment for Tiny Object Detection	https://github.com/chasel-tsui/mmdet-rfla
2022	CVPR	QueryDet: Cascaded Sparse Query for Accelerating High-Resolution Small Object Detection	https://github.com/ChenhongyiYang/QueryDet-PyTorch
2020	TSMC	A Survey of the Four Pillars for Small Object Detection: Multiscale Representation, Contextual Information, Super-Resolution, and Region Proposal

9.数据增强

Date	Pub.	Title	Code
2021	CVPR	Scale-aware Automatic Augmentation for Object Detection	https://github.com/dvlab-research/SA-AutoAug

10.开放世界目标检测

Date	Pub.	Title	Code
2021	CVPR	Towards Open World Object Detection	https://github.com/JosephKJ/OWOD

11.长尾目标检测

Date	Pub.	Title	Code
2021	ICCV	Exploring Classification Equilibrium in Long-Tailed Object Detection	https://github.com/fcjian/loce
2021	ICCV	MOSAICOS: A Simple and Effective Use of Object-Centric Images for Long-Tailed Object Detection	https://github.com/czhang0528/MosaicOS

12. 单点目标检测

Date	Pub.	Title	Code
2022	ECCV	Point-to-Box Network for Accurate Object Detection via Single Point Supervision	https://github.com/ucas-vg/P2BNet

13. 红外目标检测

Date	Pub.	Title	Code
2022	CVPR	Target-aware Dual Adversarial Learning and a Multi-scenario Multi-Modality Benchmark to Fuse Infrared and Visible for Object Detection	https://github.com/JinyuanLiu-CV/TarDAL
2022	ECCV	Multimodal Object Detection via Probabilistic Ensembling	https://github.com/Jamie725/RGBT-detection

14. 探讨类

Date	Pub.	Title	Code
2022	arXiv	Understanding CNN Fragility When Learning With Imbalanced Data	—
2022	arXiv	TSAA: A Two-Stage Anchor Assignment Method towards Anchor Drift in Crowded Object Detection	—
2019	arXiv	Empirical Upper Bound in Object Detection and More	—

15. diffusion

Date	Pub.	Title	Code
2022	arXiv	DiffusionDet: Diffusion Model for Object Detection	https://github.com/shoufachen/diffusiondet

16. Loss相关

Date	Pub.	Title	Code
2021	arXiv	Focal and Efficient IOU Loss for Accurate Bounding Box Regression
2021	arXiv	Alpha-IoU: A Family of Power Intersection over Union Losses for Bounding Box Regression	https://github.com/jacobi93/alpha-iou
2020	AAAI	Distance-IoU Loss: Faster and Better Learning for Bounding Box Regression	https://github.com/Zzh-tju/DIoU
2019	CVPR	Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regression	-

本文链接：https://blog.csdn.net/wxd1233/article/details/125902245

原作者删帖不实内容删帖广告或垃圾文章投诉

智能推荐

Kotlin 解压缩_kotlin 对上传的压缩包进行分析-程序员宅基地

文章浏览阅读638次。fun unZip(zipFile: String, context: Context) { var outputStream: OutputStream? = null var inputStream: InputStream? = null try { val zf = ZipFile(zipFile) val entries = zf.entries() while (en..._kotlin 对上传的压缩包进行分析

64K方法数限制解决办法_java函数大于64k编译失败-程序员宅基地

文章浏览阅读1.9k次。随着业务逻辑越来越多，业务模块也越来越大，不可避免会遇到64K方法数的限制。最直观的表现就是编译报错：较早版本的编译系统中，错误内容如下：Conversion to Dalvik format failed:Unable to execute dex: method ID not in [0, 0xffff]: 65536较新版本的编译系统中，错误内容如下：trouble writing outp_java函数大于64k编译失败

案例分享——低压电力线载波通信模组（借助电源线实现远距离数据传输、宽压输入、波特率范围广、应用场景多样化）_电力载波模块csdn-程序员宅基地

文章浏览阅读2k次，点赞7次，收藏10次。物联网领域，有很多数据通信场景，因为原设备整体系统结构、运行环境等方面的限制，需求在不增加通信数据线缆的情况下实现数据的远程传输，因为特殊应用场景下考虑到环境的限制，还不能使用常规的无线通信手段，所以借助电源线缆进行传输的电力线载波技术应运而生，本次博文给大家分享的就是博主完全自主研发的低压电力线载波通信模组。_电力载波模块csdn

密码学基础_密码体制的五个要素-程序员宅基地

文章浏览阅读7.4k次。密码学基本概念密码学（Cryptology）是结合数学、计算机科学、电子与通信等学科于一体的交叉学科，研究信息系统安全的科学。起源于保密通信技术。具体来讲，研究信息系统安全保密和认证的一门科学。密码编码学，通过变换消息（对信息编码）使其保密的科学和艺术密码分析学，在未知密钥的情况下从密文推_密码体制的五个要素

python支持中文路径_基于python 处理中文路径的终极解决方法-程序员宅基地

文章浏览阅读1.9k次。1 、据说python3就没有这个问题了2 、u'字符串' 代表是unicode格式的数据，路径最好写成这个格式，别直接跟字符串'字符串'这类数据相加，相加之后type就是str，这样就会存在解码失误的问题。别直接跟字符串'字符串'这类数据相加别直接跟字符串'字符串'这类数据相加别直接跟字符串'字符串'这类数据相加unicode类型别直接跟字符串'字符串'这类数据相加说四遍3 、有些读取的方式偏偏..._python 路径中文

阿里云 B 站直播首秀，用 Serverless 搭个游戏机？-程序员宅基地

文章浏览阅读107次。最近，阿云 B 站没声音，是在憋大招！8月5日周四 19:00 是阿里云的直播首秀，给大家请来了 Forrester 评分世界第一的 Serverless 团队产品经理江昱，给大家在线...._阿里云直播b站

随便推点

AS 3.1.3连续依赖多个Module，导致访问不到Module中的类_为什么as在一个包下建了多个module,缺无法打开了-程序员宅基地

文章浏览阅读1.1k次。我好苦啊，半夜还在打代码。还出bug，狗日的。问题是这样的：我在新建的项目里，建了两个Module: fiora-ec和fiora-core。项目的依赖顺序是这样的，App依赖fiora-ec,fiora-ec又依赖于fiora-core，因为这种依赖关系，所有可以在app和fiora-ec中删除一些不必要的引入，比如这个玩意儿：com.android.support:appcompat-v7:..._为什么as在一个包下建了多个module,缺无法打开了

Magento 常用插件二-程序员宅基地

文章浏览阅读1.4k次。1. SMTP 插件 URL：http://www.magentocommerce.com/magento-connect/TurboSMTP/extension/4415/aschroder_turbosmtp KEY：magento-community/Aschroder_TurboSmtp 2. Email Template Adapter..._magento extension pour ricardo.ch

【连载】【FPGA黑金开发板】Verilog HDL那些事儿--低级建模的资源（六）-程序员宅基地

文章浏览阅读161次。声明：本文为原创作品，版权归akuei2及黑金动力社区共同所有，如需转载，请注明出处http://www.cnblogs.com/kingst/ 2.5 低级建模的资源低级建模有讲求资源的分配，目的是使用“图形”来提高建模的解读性。图上是低级建模最基本的建模框图，估计大家在实验一和实验二已经眼熟过。功能模块（低级功能模块）是一个水平的长方形，而控制模块（低级控制模块）是矩形。组..._cyclone ep2c8q208c黑金开发板

R语言实用案例分析-1_r语言案例分析-程序员宅基地

文章浏览阅读2.2w次，点赞10次，收藏63次。在日常生活和实际应用当中，我们经常会用到统计方面的知识，比如求最大值，求平均值等等。R语言是一门统计学语言，他可以方便的完成统计相关的计算，下面我们就来看一个相关案例。1. 背景最近西安交大大数据专业二班，开设了Java和大数据技术课程，班级人数共100人。2. 需求通过R语言完成该100位同学学号的生成，同时使用R语言模拟生成Java和大数据技术成绩，成绩满分为100，需要满足正_r语言案例分析

Java知识体系总结（2024版），这一次带你搞懂Spring代理创建过程-程序员宅基地

文章浏览阅读639次，点赞11次，收藏26次。虽然我个人也经常自嘲，十年之后要去成为外卖专员，但实际上依靠自身的努力，是能够减少三十五岁之后的焦虑的，毕竟好的架构师并不多。架构师，是我们大部分技术人的职业目标，一名好的架构师来源于机遇(公司)、个人努力(吃得苦、肯钻研)、天分(真的热爱)的三者协作的结果，实践+机遇+努力才能助你成为优秀的架构师。如果你也想成为一名好的架构师，那或许这份Java成长笔记你需要阅读阅读，希望能够对你的职业发展有所帮助。一个人可以走的很快，但一群人才能走的更远。

车辆动力学及在Unity、UE4中的实现_unity 车辆动力学模型-程序员宅基地

文章浏览阅读3.9k次，点赞9次，收藏53次。受力分析直线行驶时的车轮受力如下：水平方向上，所受合力为：F=Ft+Fw+FfF=F_t+F_w+F_fF=Ft+Fw+Ff其中，FtF_tFt为牵引力，FwF_wFw为空气阻力，FfF_fFf为滚动阻力，下面我们将逐个介绍。驱动力先来说扭矩，扭矩是使物体发生旋转的一个特殊力矩，等于力和力臂的乘积，单位为N∙mN∙mN∙m：设驱动轴的扭矩为TtT_tTt，车轮半径为rrr，那么牵引力：Ft=Tt⁄rF_t=T_t⁄rFt=Tt⁄r如何求得驱动轴扭矩TtT_tTt呢？_unity 车辆动力学模型