[
  {
    "path": "Date.md",
    "content": "2019-10-11 Update 1 Project\n\n**Detectron2**\n\n- intro: Detectron2 is FAIR's next-generation research platform for object detection and segmentation.\n- [blog](https://ai.facebook.com/blog/-detectron2-a-pytorch-based-modular-object-detection-library-/)\n- code: <https://github.com/facebookresearch/detectron2>\n\n2019-09-06 Update 1 paper\n\n**Imbalance Problems in Object Detection: A Review**\n\n- intro: under review at TPAMI\n- arXiv: <https://arxiv.org/abs/1909.00169>\n\n2019-08-14 Update 1 paper\n\n**Recent Advances in Deep Learning for Object Detection**\n\n- intro: From 2013 (OverFeat) to 2019 (DetNAS)\n- arXiv: <https://arxiv.org/abs/1908.03673>\n\n2019-07-24 Update 1 paper\n\n**A Survey of Deep Learning-based Object Detection**\n\n- intro：From Fast R-CNN to NAS-FPN\n\n- arXiv：<https://arxiv.org/abs/1907.09408>\n\n2019-05-17 Update 1 paper\n\n**Object Detection in 20 Years: A Survey**\n\n- intro：This work has been submitted to the IEEE TPAMI for possible publication\n- arXiv：<https://arxiv.org/abs/1905.05055>\n\n2019-04-05 Update 1 paper\n\n**Comparison Network for One-Shot Conditional Object Detection**\n\n- arXiv: https://arxiv.org/abs/1904.02317\n\n2019-03-05 Update 1 paper\n\n**Feature Selective Anchor-Free Module for Single-Shot Object Detection**\n\n- intro: CVPR 2019\n- arXiv: https://arxiv.org/abs/1903.00621\n\n2019-02-15 Update 3 detection toolbox\n\n- [Detectron(FAIR)](https://github.com/facebookresearch/Detectron): Detectron is Facebook AI Research's software system that implements state-of-the-art object detection algorithms, including [Mask R-CNN](https://arxiv.org/abs/1703.06870). It is written in Python and powered by the [Caffe2](https://github.com/caffe2/caffe2) deep learning framework.\n\n- [maskrcnn-benchmark(FAIR)](https://github.com/facebookresearch/maskrcnn-benchmark): Fast, modular reference implementation of Instance Segmentation and Object Detection algorithms in PyTorch.\n\n- [mmdetection(SenseTime&CUHK)](https://github.com/open-mmlab/mmdetection): mmdetection is an open source object detection toolbox based on PyTorch. It is a part of the open-mmlab project developed by [Multimedia Laboratory, CUHK](http://mmlab.ie.cuhk.edu.hk/).\n\n2019-01-25 Update 5 papers\n\n**3D Backbone Network for 3D Object Detection**\n\n- arXiv: https://arxiv.org/abs/1901.08373\n\n**Object Detection based on Region Decomposition and Assembly**\n\n- intro: AAAI 2019\n\n- arXiv: https://arxiv.org/abs/1901.08225\n\n**Bottom-up Object Detection by Grouping Extreme and Center Points**\n\n- intro: one stage 43.2% on COCO test-dev\n- arXiv: https://arxiv.org/abs/1901.08043\n- github: https://github.com/xingyizhou/ExtremeNet\n\n**ORSIm Detector: A Novel Object Detection Framework in Optical Remote Sensing Imagery Using Spatial-Frequency Channel Features**\n\n- intro: IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING\n\n- arXiv: https://arxiv.org/abs/1901.07925\n\n**Consistent Optimization for Single-Shot Object Detection**\n\n- intro: improves RetinaNet from 39.1 AP to 40.1 AP on COCO datase\n\n- arXiv: https://arxiv.org/abs/1901.06563\n\n2019-01-15 Update 1 paper\n\n**Learning Pairwise Relationship for Multi-object Detection in Crowded Scenes**\n\n- arXiv: https://arxiv.org/abs/1901.03796\n\n2019-01-14 Update 1 paper\n\n**RetinaMask: Learning to predict masks improves state-of-the-art single-shot detection for free**\n\n- arXiv: https://arxiv.org/abs/1901.03353\n- github: https://github.com/chengyangfu/retinamask\n\n2019-01-12 Update 1 paper\n\n**Region Proposal by Guided Anchoring**\n\n- intro: CUHK - SenseTime Joint Lab\n- arXiv: https://arxiv.org/abs/1901.03278\n\n2019-01-08 Update 1 paper\n\n**Scale-Aware Trident Networks for Object Detection**\n\n- intro: mAP of **48.4** on the COCO dataset\n- arXiv: https://arxiv.org/abs/1901.01892\n\n2019-01-04 Update 1 paper\n\n**Large-Scale Object Detection of Images from Network Cameras in Variable Ambient Lighting Conditions**\n\n- arXiv: https://arxiv.org/abs/1812.11901\n\n2018-12-13 Update 1 paper\n\n**Strong-Weak Distribution Alignment for Adaptive Object Detection**\n\n- arXiv: https://arxiv.org/abs/1812.04798\n\n2018-12-05 Update 3 papers\n\n**AutoFocus: Efficient Multi-Scale Inference**\n\n- intro: AutoFocus obtains an **mAP of 47.9%** (68.3% at 50% overlap) on the **COCO test-dev** set while processing **6.4 images per second on a Titan X (Pascal) GPU** \n- arXiv: https://arxiv.org/abs/1812.01600\n\n**NOTE-RCNN: NOise Tolerant Ensemble RCNN for Semi-Supervised Object Detection**\n\n- intro: Google Could\n- arXiv: https://arxiv.org/abs/1812.00124\n\n**SPLAT: Semantic Pixel-Level Adaptation Transforms for Detection**\n\n- intro: UC Berkeley\n- arXiv: https://arxiv.org/abs/1812.00929\n\n2018-12-04 Update 10 papers\n\n**Grid R-CNN**\n\n- intro: SenseTime\n- arXiv: https://arxiv.org/abs/1811.12030\n\n**Deformable ConvNets v2: More Deformable, Better Results**\n\n- intro: Microsoft Research Asia\n\n- arXiv: https://arxiv.org/abs/1811.11168\n\n**Anchor Box Optimization for Object Detection**\n\n- intro: Microsoft Research\n- arXiv: https://arxiv.org/abs/1812.00469\n\n**Efficient Coarse-to-Fine Non-Local Module for the Detection of Small Objects**\n\n- intro: https://arxiv.org/abs/1811.12152\n\n**NOTE-RCNN: NOise Tolerant Ensemble RCNN for Semi-Supervised Object Detection**\n\n- arXiv: https://arxiv.org/abs/1812.00124\n\n**Learning RoI Transformer for Detecting Oriented Objects in Aerial Images**\n\n- arXiv: https://arxiv.org/abs/1812.00155\n\n**Integrated Object Detection and Tracking with Tracklet-Conditioned Detection**\n\n- intro: Microsoft Research Asia\n- arXiv: https://arxiv.org/abs/1811.11167\n\n**Deep Regionlets: Blended Representation and Deep Learning for Generic Object Detection**\n\n- arXiv: https://arxiv.org/abs/1811.11318\n\n **Gradient Harmonized Single-stage Detector**\n\n- intro: AAAI 2019\n- arXiv: https://arxiv.org/abs/1811.05181\n\n**CFENet: Object Detection with Comprehensive Feature Enhancement Module**\n\n- intro: ACCV 2018\n- github: https://github.com/qijiezhao/CFENet\n\n2018-11-19\n\n**DeRPN: Taking a further step toward more general object detection**\n\n- intro: AAAI 2019\n- arXiv: https://arxiv.org/abs/1811.06700\n- github: https://github.com/HCIILAB/DeRPN\n\n2018-11-14\n\n**M2Det: A Single-Shot Object Detector based on Multi-Level Feature Pyramid Network**\n\n- intro: AAAI 2019\n- arXiv: https://arxiv.org/abs/1811.04533\n- github: https://github.com/qijiezhao/M2Det\n\n2018-10-31\n\n**Hybrid Knowledge Routed Modules for Large-scale Object Detection**\n\n- intro: Sun Yat-Sen University & Huawei Noah’s Ark Lab\n- arXiv: https://arxiv.org/abs/1810.12681\n- github: https://github.com/chanyn/HKRM\n\n2018-10-08\n\n**Weakly Supervised Object Detection in Artworks**\n\n- intro: ECCV 2018 Workshop Computer Vision for Art Analysis\n- arXiv: https://arxiv.org/abs/1810.02569\n- Datasets: https://wsoda.telecom-paristech.fr/downloads/dataset/IconArt_v1.zip\n\n**Cross-Domain Weakly-Supervised Object Detection through Progressive Domain Adaptation**\n\n- intro: CVPR 2018\n- arXiv: https://arxiv.org/abs/1803.11365\n- homepage: https://naoto0804.github.io/cross_domain_detection/\n- paper: http://openaccess.thecvf.com/content_cvpr_2018/html/Inoue_Cross-Domain_Weakly-Supervised_Object_CVPR_2018_paper.html\n- github: https://github.com/naoto0804/cross-domain-detection\n\n2018-09-26\n\n**Object Detection from Scratch with Deep Supervision**\n\n- intro: This is an extended version of DSOD\n- arXiv: https://arxiv.org/abs/1809.09294\n\n2018-09-25\n\n**《Softer-NMS: Rethinking Bounding Box Regression for Accurate Object Detection》**\n\n- intro: CMU & Face++\n- arXiv: https://arxiv.org/abs/1809.08545\n- github: https://github.com/yihui-he/softer-NMS\n\n2018-09-21\n\n**《Receptive Field Block Net for Accurate and Fast Object Detection》**\n\n- intro: ECCV 2018\n- arXiv: [https://arxiv.org/abs/1711.07767](https://arxiv.org/abs/1711.07767)\n- github: [https://github.com/ruinmessi/RFBNet](https://github.com/ruinmessi/RFBNet)\n\n2018-09-11\n\n**《Recent Advances in Object Detection in the Age of Deep Convolutional Neural Networks》**\n\n- intro: awesome\n\n\n- arXiv: https://arxiv.org/abs/1809.03193\n\n2018-09-10\n\n**《Deep Learning for Generic Object Detection: A Survey》**\n\n- intro: Submitted to IJCV 2018\n- arXiv: https://arxiv.org/abs/1809.02165\n\n2018-08-27\n\n**Deep Feature Pyramid Reconfiguration for Object Detection**\n\n- intro: ECCV 2018\n- arXiv: https://arxiv.org/abs/1808.07993\n\n2018-08-17\n\n**R3-Net: A Deep Network for Multi-oriented Vehicle Detection in Aerial Images and Videos**\n\n- arxiv: https://arxiv.org/abs/1808.05560\n- youtube: https://youtu.be/xCYD-tYudN0\n\n2018-08-14\n\n**《Unsupervised Hard Example Mining from Videos for Improved Object Detection》**\n\n- intro: ECCV 2018\n- arXiv: https://arxiv.org/abs/1808.04285\n\n2018-08-10\n\n**CornerNet: Detecting Objects as Paired Keypoints**\n\n- intro: ECCV 2018\n- arXiv: https://arxiv.org/abs/1808.01244\n\n2018-07-30\n\n**Acquisition of Localization Confidence for Accurate Object Detection**\n\n- intro: ECCV 2018\n- arXiv: https://arxiv.org/abs/1807.11590\n- github: https://github.com/vacancy/PreciseRoIPooling"
  },
  {
    "path": "README.md",
    "content": "# object-detection\r\n\r\n[TOC]\r\n\r\nThis is a list of awesome articles about object detection. If you want to read the paper according to time, you can refer to [Date](Date.md).\r\n\r\n- R-CNN\r\n- Fast R-CNN\r\n- Faster R-CNN\r\n- Mask R-CNN\r\n- Light-Head R-CNN\r\n- Cascade R-CNN\r\n- SPP-Net\r\n- YOLO\r\n- YOLOv2\r\n- YOLOv3\r\n- YOLT\r\n- SSD\r\n- DSSD\r\n- FSSD\r\n- ESSD\r\n- MDSSD\r\n- Pelee\r\n- Fire SSD\r\n- R-FCN\r\n- FPN\r\n- DSOD\r\n- RetinaNet\r\n- MegDet\r\n- RefineNet\r\n- DetNet\r\n- SSOD\r\n- CornerNet\r\n- M2Det\r\n- 3D Object Detection\r\n- ZSD（Zero-Shot Object Detection）\r\n- OSD（One-Shot object Detection）\r\n- Weakly Supervised Object Detection\r\n- Softer-NMS\r\n- 2018\r\n- 2019\r\n- Other\r\n\r\nBased on handong1587's github: https://handong1587.github.io/deep_learning/2015/10/09/object-detection.html\r\n\r\n# Survey\r\n\r\n**Imbalance Problems in Object Detection: A Review**\r\n\r\n- intro: under review at TPAMI\r\n- arXiv: <https://arxiv.org/abs/1909.00169>\r\n\r\n**Recent Advances in Deep Learning for Object Detection**\r\n\r\n- intro: From 2013 (OverFeat) to 2019 (DetNAS)\r\n- arXiv: <https://arxiv.org/abs/1908.03673>\r\n\r\n**A Survey of Deep Learning-based Object Detection**\r\n\r\n- intro：From Fast R-CNN to NAS-FPN\r\n\r\n- arXiv：<https://arxiv.org/abs/1907.09408>\r\n\r\n**Object Detection in 20 Years: A Survey**\r\n\r\n- intro：This work has been submitted to the IEEE TPAMI for possible publication\r\n- arXiv：<https://arxiv.org/abs/1905.05055>\r\n\r\n**《Recent Advances in Object Detection in the Age of Deep Convolutional Neural Networks》**\r\n\r\n- intro: awesome\r\n\r\n\r\n- arXiv: https://arxiv.org/abs/1809.03193\r\n\r\n**《Deep Learning for Generic Object Detection: A Survey》**\r\n\r\n- intro: Submitted to IJCV 2018\r\n- arXiv: https://arxiv.org/abs/1809.02165\r\n\r\n# Papers&Codes\r\n\r\n## R-CNN\r\n\r\n**Rich feature hierarchies for accurate object detection and semantic segmentation**\r\n\r\n- intro: R-CNN\r\n- arxiv: <http://arxiv.org/abs/1311.2524>\r\n- supp: <http://people.eecs.berkeley.edu/~rbg/papers/r-cnn-cvpr-supp.pdf>\r\n- slides: <http://www.image-net.org/challenges/LSVRC/2013/slides/r-cnn-ilsvrc2013-workshop.pdf>\r\n- slides: <http://www.cs.berkeley.edu/~rbg/slides/rcnn-cvpr14-slides.pdf>\r\n- github: <https://github.com/rbgirshick/rcnn>\r\n- notes: <http://zhangliliang.com/2014/07/23/paper-note-rcnn/>\r\n- caffe-pr(\"Make R-CNN the Caffe detection example\"): <https://github.com/BVLC/caffe/pull/482>\r\n\r\n## Fast R-CNN\r\n\r\n**Fast R-CNN**\r\n\r\n- arxiv: <http://arxiv.org/abs/1504.08083>\r\n- slides: <http://tutorial.caffe.berkeleyvision.org/caffe-cvpr15-detection.pdf>\r\n- github: <https://github.com/rbgirshick/fast-rcnn>\r\n- github(COCO-branch): <https://github.com/rbgirshick/fast-rcnn/tree/coco>\r\n- webcam demo: <https://github.com/rbgirshick/fast-rcnn/pull/29>\r\n- notes: <http://zhangliliang.com/2015/05/17/paper-note-fast-rcnn/>\r\n- notes: <http://blog.csdn.net/linj_m/article/details/48930179>\r\n- github(\"Fast R-CNN in MXNet\"): <https://github.com/precedenceguo/mx-rcnn>\r\n- github: <https://github.com/mahyarnajibi/fast-rcnn-torch>\r\n- github: <https://github.com/apple2373/chainer-simple-fast-rnn>\r\n- github: <https://github.com/zplizzi/tensorflow-fast-rcnn>\r\n\r\n**A-Fast-RCNN: Hard Positive Generation via Adversary for Object Detection**\r\n\r\n- intro: CVPR 2017\r\n- arxiv: <https://arxiv.org/abs/1704.03414>\r\n- paper: <http://abhinavsh.info/papers/pdfs/adversarial_object_detection.pdf>\r\n- github(Caffe): <https://github.com/xiaolonw/adversarial-frcnn>\r\n\r\n## Faster R-CNN\r\n\r\n**Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks**\r\n\r\n- intro: NIPS 2015\r\n- arxiv: <http://arxiv.org/abs/1506.01497>\r\n- gitxiv: <http://www.gitxiv.com/posts/8pfpcvefDYn2gSgXk/faster-r-cnn-towards-real-time-object-detection-with-region>\r\n- slides: <http://web.cs.hacettepe.edu.tr/~aykut/classes/spring2016/bil722/slides/w05-FasterR-CNN.pdf>\r\n- github(official, Matlab): <https://github.com/ShaoqingRen/faster_rcnn>\r\n- github(Caffe): <https://github.com/rbgirshick/py-faster-rcnn>\r\n- github(MXNet): <https://github.com/msracver/Deformable-ConvNets/tree/master/faster_rcnn>\r\n- github(PyTorch--recommend): <https://github.com//jwyang/faster-rcnn.pytorch>\r\n- github: <https://github.com/mitmul/chainer-faster-rcnn>\r\n- github(Torch):: <https://github.com/andreaskoepf/faster-rcnn.torch>\r\n- github(Torch):: <https://github.com/ruotianluo/Faster-RCNN-Densecap-torch>\r\n- github(TensorFlow): <https://github.com/smallcorgi/Faster-RCNN_TF>\r\n- github(TensorFlow): <https://github.com/CharlesShang/TFFRCNN>\r\n- github(C++ demo): <https://github.com/YihangLou/FasterRCNN-Encapsulation-Cplusplus>\r\n- github(Keras): <https://github.com/yhenon/keras-frcnn>\r\n- github: <https://github.com/Eniac-Xie/faster-rcnn-resnet>\r\n- github(C++): <https://github.com/D-X-Y/caffe-faster-rcnn/tree/dev>\r\n\r\n**R-CNN minus R**\r\n\r\n- intro: BMVC 2015\r\n- arxiv: <http://arxiv.org/abs/1506.06981>\r\n\r\n**Faster R-CNN in MXNet with distributed implementation and data parallelization**\r\n\r\n- github: <https://github.com/dmlc/mxnet/tree/master/example/rcnn>\r\n\r\n**Contextual Priming and Feedback for Faster R-CNN**\r\n\r\n- intro: ECCV 2016. Carnegie Mellon University\r\n- paper: <http://abhinavsh.info/context_priming_feedback.pdf>\r\n- poster: <http://www.eccv2016.org/files/posters/P-1A-20.pdf>\r\n\r\n**An Implementation of Faster RCNN with Study for Region Sampling**\r\n\r\n- intro: Technical Report, 3 pages. CMU\r\n- arxiv: <https://arxiv.org/abs/1702.02138>\r\n- github: <https://github.com/endernewton/tf-faster-rcnn>\r\n- github: https://github.com/ruotianluo/pytorch-faster-rcnn\r\n\r\n**Interpretable R-CNN**\r\n\r\n- intro: North Carolina State University & Alibaba\r\n- keywords: AND-OR Graph (AOG)\r\n- arxiv: <https://arxiv.org/abs/1711.05226>\r\n\r\n**Domain Adaptive Faster R-CNN for Object Detection in the Wild**\r\n\r\n- intro: CVPR 2018. ETH Zurich & ESAT/PSI\r\n- arxiv: <https://arxiv.org/abs/1803.03243>\r\n\r\n## Mask R-CNN\r\n\r\n- arxiv: <http://arxiv.org/abs/1703.06870>\r\n- github(Keras): https://github.com/matterport/Mask_RCNN\r\n- github(Caffe2): https://github.com/facebookresearch/Detectron\r\n- github(Pytorch): <https://github.com/wannabeOG/Mask-RCNN>\r\n- github(MXNet): https://github.com/TuSimple/mx-maskrcnn\r\n- github(Chainer): https://github.com/DeNA/Chainer_Mask_R-CNN\r\n\r\n## Light-Head R-CNN\r\n\r\n**Light-Head R-CNN: In Defense of Two-Stage Object Detector**\r\n\r\n- intro: Tsinghua University & Megvii Inc\r\n- arxiv: <https://arxiv.org/abs/1711.07264>\r\n- github(offical): https://github.com/zengarden/light_head_rcnn\r\n- github: <https://github.com/terrychenism/Deformable-ConvNets/blob/master/rfcn/symbols/resnet_v1_101_rfcn_light.py#L784>\r\n\r\n## Cascade R-CNN\r\n\r\n**Cascade R-CNN: Delving into High Quality Object Detection**\r\n\r\n- arxiv: <https://arxiv.org/abs/1712.00726>\r\n- github: <https://github.com/zhaoweicai/cascade-rcnn>\r\n\r\n## SPP-Net\r\n\r\n**Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition**\r\n\r\n- intro: ECCV 2014 / TPAMI 2015\r\n- arxiv: <http://arxiv.org/abs/1406.4729>\r\n- github: <https://github.com/ShaoqingRen/SPP_net>\r\n- notes: <http://zhangliliang.com/2014/09/13/paper-note-sppnet/>\r\n\r\n**DeepID-Net: Deformable Deep Convolutional Neural Networks for Object Detection**\r\n\r\n- intro: PAMI 2016\r\n- intro: an extension of R-CNN. box pre-training, cascade on region proposals, deformation layers and context representations\r\n- project page: <http://www.ee.cuhk.edu.hk/%CB%9Cwlouyang/projects/imagenetDeepId/index.html>\r\n- arxiv: <http://arxiv.org/abs/1412.5661>\r\n\r\n**Object Detectors Emerge in Deep Scene CNNs**\r\n\r\n- intro: ICLR 2015\r\n- arxiv: <http://arxiv.org/abs/1412.6856>\r\n- paper: <https://www.robots.ox.ac.uk/~vgg/rg/papers/zhou_iclr15.pdf>\r\n- paper: <https://people.csail.mit.edu/khosla/papers/iclr2015_zhou.pdf>\r\n- slides: <http://places.csail.mit.edu/slide_iclr2015.pdf>\r\n\r\n**segDeepM: Exploiting Segmentation and Context in Deep Neural Networks for Object Detection**\r\n\r\n- intro: CVPR 2015\r\n- project(code+data): <https://www.cs.toronto.edu/~yukun/segdeepm.html>\r\n- arxiv: <https://arxiv.org/abs/1502.04275>\r\n- github: <https://github.com/YknZhu/segDeepM>\r\n\r\n**Object Detection Networks on Convolutional Feature Maps**\r\n\r\n- intro: TPAMI 2015\r\n- keywords: NoC\r\n- arxiv: <http://arxiv.org/abs/1504.06066>\r\n\r\n**Improving Object Detection with Deep Convolutional Networks via Bayesian Optimization and Structured Prediction**\r\n\r\n- arxiv: <http://arxiv.org/abs/1504.03293>\r\n- slides: <http://www.ytzhang.net/files/publications/2015-cvpr-det-slides.pdf>\r\n- github: <https://github.com/YutingZhang/fgs-obj>\r\n\r\n**DeepBox: Learning Objectness with Convolutional Networks**\r\n\r\n- keywords: DeepBox\r\n- arxiv: <http://arxiv.org/abs/1505.02146>\r\n- github: <https://github.com/weichengkuo/DeepBox>\r\n\r\n## YOLO\r\n\r\n**You Only Look Once: Unified, Real-Time Object Detection**\r\n\r\n[![img](https://camo.githubusercontent.com/e69d4118b20a42de4e23b9549f9a6ec6dbbb0814/687474703a2f2f706a7265646469652e636f6d2f6d656469612f66696c65732f6461726b6e65742d626c61636b2d736d616c6c2e706e67)](https://camo.githubusercontent.com/e69d4118b20a42de4e23b9549f9a6ec6dbbb0814/687474703a2f2f706a7265646469652e636f6d2f6d656469612f66696c65732f6461726b6e65742d626c61636b2d736d616c6c2e706e67)\r\n\r\n- arxiv: <http://arxiv.org/abs/1506.02640>\r\n- code: <https://pjreddie.com/darknet/yolov1/>\r\n- github: <https://github.com/pjreddie/darknet>\r\n- blog: <https://pjreddie.com/darknet/yolov1/>\r\n- slides: <https://docs.google.com/presentation/d/1aeRvtKG21KHdD5lg6Hgyhx5rPq_ZOsGjG5rJ1HP7BbA/pub?start=false&loop=false&delayms=3000&slide=id.p>\r\n- reddit: <https://www.reddit.com/r/MachineLearning/comments/3a3m0o/realtime_object_detection_with_yolo/>\r\n- github: <https://github.com/gliese581gg/YOLO_tensorflow>\r\n- github: <https://github.com/xingwangsfu/caffe-yolo>\r\n- github: <https://github.com/frankzhangrui/Darknet-Yolo>\r\n- github: <https://github.com/BriSkyHekun/py-darknet-yolo>\r\n- github: <https://github.com/tommy-qichang/yolo.torch>\r\n- github: <https://github.com/frischzenger/yolo-windows>\r\n- github: <https://github.com/AlexeyAB/yolo-windows>\r\n- github: <https://github.com/nilboy/tensorflow-yolo>\r\n\r\n**darkflow - translate darknet to tensorflow. Load trained weights, retrain/fine-tune them using tensorflow, export constant graph def to C++**\r\n\r\n- blog: <https://thtrieu.github.io/notes/yolo-tensorflow-graph-buffer-cpp>\r\n- github: <https://github.com/thtrieu/darkflow>\r\n\r\n**Start Training YOLO with Our Own Data**\r\n\r\n[![img](https://camo.githubusercontent.com/2f99b692dd7ce47d7832385f3e8a6654e680d92a/687474703a2f2f6775616e6768616e2e696e666f2f626c6f672f656e2f77702d636f6e74656e742f75706c6f6164732f323031352f31322f696d616765732d34302e6a7067)](https://camo.githubusercontent.com/2f99b692dd7ce47d7832385f3e8a6654e680d92a/687474703a2f2f6775616e6768616e2e696e666f2f626c6f672f656e2f77702d636f6e74656e742f75706c6f6164732f323031352f31322f696d616765732d34302e6a7067)\r\n\r\n- intro: train with customized data and class numbers/labels. Linux / Windows version for darknet.\r\n- blog: <http://guanghan.info/blog/en/my-works/train-yolo/>\r\n- github: <https://github.com/Guanghan/darknet>\r\n\r\n**YOLO: Core ML versus MPSNNGraph**\r\n\r\n- intro: Tiny YOLO for iOS implemented using CoreML but also using the new MPS graph API.\r\n- blog: <http://machinethink.net/blog/yolo-coreml-versus-mps-graph/>\r\n- github: <https://github.com/hollance/YOLO-CoreML-MPSNNGraph>\r\n\r\n**TensorFlow YOLO object detection on Android**\r\n\r\n- intro: Real-time object detection on Android using the YOLO network with TensorFlow\r\n- github: <https://github.com/natanielruiz/android-yolo>\r\n\r\n**Computer Vision in iOS – Object Detection**\r\n\r\n- blog: <https://sriraghu.com/2017/07/12/computer-vision-in-ios-object-detection/>\r\n- github:<https://github.com/r4ghu/iOS-CoreML-Yolo>\r\n\r\n## YOLOv2\r\n\r\n**YOLO9000: Better, Faster, Stronger**\r\n\r\n- arxiv: <https://arxiv.org/abs/1612.08242>\r\n- code: <http://pjreddie.com/yolo9000/>    https://pjreddie.com/darknet/yolov2/\r\n- github(Chainer): <https://github.com/leetenki/YOLOv2>\r\n- github(Keras): <https://github.com/allanzelener/YAD2K>\r\n- github(PyTorch): <https://github.com/longcw/yolo2-pytorch>\r\n- github(Tensorflow): <https://github.com/hizhangp/yolo_tensorflow>\r\n- github(Windows): <https://github.com/AlexeyAB/darknet>\r\n- github: <https://github.com/choasUp/caffe-yolo9000>\r\n- github: <https://github.com/philipperemy/yolo-9000>\r\n- github(TensorFlow): <https://github.com/KOD-Chen/YOLOv2-Tensorflow>\r\n- github(Keras): <https://github.com/yhcc/yolo2>\r\n- github(Keras): <https://github.com/experiencor/keras-yolo2>\r\n- github(TensorFlow): <https://github.com/WojciechMormul/yolo2>\r\n\r\n**darknet_scripts**\r\n\r\n- intro: Auxilary scripts to work with (YOLO) darknet deep learning famework. AKA -> How to generate YOLO anchors?\r\n- github: <https://github.com/Jumabek/darknet_scripts>\r\n\r\n**Yolo_mark: GUI for marking bounded boxes of objects in images for training Yolo v2**\r\n\r\n- github: <https://github.com/AlexeyAB/Yolo_mark>\r\n\r\n**LightNet: Bringing pjreddie's DarkNet out of the shadows**\r\n\r\n<https://github.com//explosion/lightnet>\r\n\r\n**YOLO v2 Bounding Box Tool**\r\n\r\n- intro: Bounding box labeler tool to generate the training data in the format YOLO v2 requires.\r\n- github: <https://github.com/Cartucho/yolo-boundingbox-labeler-GUI>\r\n\r\n**Loss Rank Mining: A General Hard Example Mining Method for Real-time Detectors**\r\n\r\n- intro: **LRM** is the first hard example mining strategy which could fit YOLOv2 perfectly and make it better applied in series of real scenarios where both real-time rates and accurate detection are strongly demanded.\r\n- arxiv: https://arxiv.org/abs/1804.04606\r\n\r\n**Object detection at 200 Frames Per Second**\r\n\r\n- intro: faster than Tiny-Yolo-v2\r\n- arxiv: https://arxiv.org/abs/1805.06361\r\n\r\n**Event-based Convolutional Networks for Object Detection in Neuromorphic Cameras**\r\n\r\n- intro: YOLE--Object Detection in Neuromorphic Cameras\r\n- arxiv:https://arxiv.org/abs/1805.07931\r\n\r\n**OmniDetector: With Neural Networks to Bounding Boxes**\r\n\r\n- intro: a person detector on n fish-eye images of indoor scenes（NIPS 2018）\r\n- arxiv:https://arxiv.org/abs/1805.08503\r\n- datasets:https://gitlab.com/omnidetector/omnidetector\r\n\r\n## YOLOv3\r\n\r\n**YOLOv3: An Incremental Improvement**\r\n\r\n- arxiv:https://arxiv.org/abs/1804.02767\r\n- paper:https://pjreddie.com/media/files/papers/YOLOv3.pdf\r\n- code: <https://pjreddie.com/darknet/yolo/>\r\n- github(Official):https://github.com/pjreddie/darknet\r\n- github:https://github.com/mystic123/tensorflow-yolo-v3\r\n- github:https://github.com/experiencor/keras-yolo3\r\n- github:https://github.com/qqwweee/keras-yolo3\r\n- github:https://github.com/marvis/pytorch-yolo3\r\n- github:https://github.com/ayooshkathuria/pytorch-yolo-v3\r\n- github:https://github.com/ayooshkathuria/YOLO_v3_tutorial_from_scratch\r\n- github:https://github.com/eriklindernoren/PyTorch-YOLOv3\r\n- github:https://github.com/ultralytics/yolov3\r\n- github:https://github.com/BobLiu20/YOLOv3_PyTorch\r\n- github:https://github.com/andy-yun/pytorch-0.4-yolov3\r\n- github:https://github.com/DeNA/PyTorch_YOLOv3\r\n\r\n## YOLT\r\n\r\n**You Only Look Twice: Rapid Multi-Scale Object Detection In Satellite Imagery**\r\n\r\n- intro: Small Object Detection\r\n\r\n\r\n- arxiv:https://arxiv.org/abs/1805.09512\r\n- github:https://github.com/avanetten/yolt\r\n\r\n## SSD\r\n\r\n**SSD: Single Shot MultiBox Detector**\r\n\r\n[![img](https://camo.githubusercontent.com/ad9b147ed3a5f48ffb7c3540711c15aa04ce49c6/687474703a2f2f7777772e63732e756e632e6564752f7e776c69752f7061706572732f7373642e706e67)](https://camo.githubusercontent.com/ad9b147ed3a5f48ffb7c3540711c15aa04ce49c6/687474703a2f2f7777772e63732e756e632e6564752f7e776c69752f7061706572732f7373642e706e67)\r\n\r\n- intro: ECCV 2016 Oral\r\n- arxiv: <http://arxiv.org/abs/1512.02325>\r\n- paper: <http://www.cs.unc.edu/~wliu/papers/ssd.pdf>\r\n- slides: [http://www.cs.unc.edu/%7Ewliu/papers/ssd_eccv2016_slide.pdf](http://www.cs.unc.edu/~wliu/papers/ssd_eccv2016_slide.pdf)\r\n- github(Official): <https://github.com/weiliu89/caffe/tree/ssd>\r\n- video: <http://weibo.com/p/2304447a2326da963254c963c97fb05dd3a973>\r\n- github: <https://github.com/zhreshold/mxnet-ssd>\r\n- github: <https://github.com/zhreshold/mxnet-ssd.cpp>\r\n- github: <https://github.com/rykov8/ssd_keras>\r\n- github: <https://github.com/balancap/SSD-Tensorflow>\r\n- github: <https://github.com/amdegroot/ssd.pytorch>\r\n- github(Caffe): <https://github.com/chuanqi305/MobileNet-SSD>\r\n\r\n**What's the diffience in performance between this new code you pushed and the previous code? #327**\r\n\r\n<https://github.com/weiliu89/caffe/issues/327>\r\n\r\n## DSSD\r\n\r\n**DSSD : Deconvolutional Single Shot Detector**\r\n\r\n- intro: UNC Chapel Hill & Amazon Inc\r\n- arxiv: <https://arxiv.org/abs/1701.06659>\r\n- github: <https://github.com/chengyangfu/caffe/tree/dssd>\r\n- github: <https://github.com/MTCloudVision/mxnet-dssd>\r\n- demo: <http://120.52.72.53/www.cs.unc.edu/c3pr90ntc0td/~cyfu/dssd_lalaland.mp4>\r\n\r\n**Enhancement of SSD by concatenating feature maps for object detection**\r\n\r\n- intro: rainbow SSD (R-SSD)\r\n- arxiv: <https://arxiv.org/abs/1705.09587>\r\n\r\n**Context-aware Single-Shot Detector**\r\n\r\n- keywords: CSSD, DiCSSD, DeCSSD, effective receptive fields (ERFs), theoretical receptive fields (TRFs)\r\n- arxiv: <https://arxiv.org/abs/1707.08682>\r\n\r\n**Feature-Fused SSD: Fast Detection for Small Objects**\r\n\r\n<https://arxiv.org/abs/1709.05054>\r\n\r\n## FSSD\r\n\r\n**FSSD: Feature Fusion Single Shot Multibox Detector**\r\n\r\n<https://arxiv.org/abs/1712.00960>\r\n\r\n**Weaving Multi-scale Context for Single Shot Detector**\r\n\r\n- intro: WeaveNet\r\n- keywords: fuse multi-scale information\r\n- arxiv: <https://arxiv.org/abs/1712.03149>\r\n\r\n## ESSD\r\n\r\n**Extend the shallow part of Single Shot MultiBox Detector via Convolutional Neural Network**\r\n\r\n<https://arxiv.org/abs/1801.05918>\r\n\r\n**Tiny SSD: A Tiny Single-shot Detection Deep Convolutional Neural Network for Real-time Embedded Object Detection**\r\n\r\n<https://arxiv.org/abs/1802.06488>\r\n\r\n## MDSSD\r\n\r\n**MDSSD: Multi-scale Deconvolutional Single Shot Detector for small objects**\r\n\r\n- arxiv: https://arxiv.org/abs/1805.07009\r\n\r\n## Pelee\r\n\r\n**Pelee: A Real-Time Object Detection System on Mobile Devices**\r\n\r\nhttps://github.com/Robert-JunWang/Pelee\r\n\r\n- intro: (ICLR 2018 workshop track)\r\n\r\n\r\n- arxiv: https://arxiv.org/abs/1804.06882\r\n- github: https://github.com/Robert-JunWang/Pelee\r\n\r\n## Fire SSD\r\n\r\n**Fire SSD: Wide Fire Modules based Single Shot Detector on Edge Device**\r\n\r\n- intro:low cost, fast speed and high mAP on  factor edge computing devices\r\n\r\n\r\n- arxiv:https://arxiv.org/abs/1806.05363\r\n\r\n## R-FCN\r\n\r\n**R-FCN: Object Detection via Region-based Fully Convolutional Networks**\r\n\r\n- arxiv: <http://arxiv.org/abs/1605.06409>\r\n- github: <https://github.com/daijifeng001/R-FCN>\r\n- github(MXNet): <https://github.com/msracver/Deformable-ConvNets/tree/master/rfcn>\r\n- github: <https://github.com/Orpine/py-R-FCN>\r\n- github: <https://github.com/PureDiors/pytorch_RFCN>\r\n- github: <https://github.com/bharatsingh430/py-R-FCN-multiGPU>\r\n- github: <https://github.com/xdever/RFCN-tensorflow>\r\n\r\n**R-FCN-3000 at 30fps: Decoupling Detection and Classification**\r\n\r\n<https://arxiv.org/abs/1712.01802>\r\n\r\n**Recycle deep features for better object detection**\r\n\r\n- arxiv: <http://arxiv.org/abs/1607.05066>\r\n\r\n## FPN\r\n\r\n**Feature Pyramid Networks for Object Detection**\r\n\r\n- intro: Facebook AI Research\r\n- arxiv: <https://arxiv.org/abs/1612.03144>\r\n\r\n**Action-Driven Object Detection with Top-Down Visual Attentions**\r\n\r\n- arxiv: <https://arxiv.org/abs/1612.06704>\r\n\r\n**Beyond Skip Connections: Top-Down Modulation for Object Detection**\r\n\r\n- intro: CMU & UC Berkeley & Google Research\r\n- arxiv: <https://arxiv.org/abs/1612.06851>\r\n\r\n**Wide-Residual-Inception Networks for Real-time Object Detection**\r\n\r\n- intro: Inha University\r\n- arxiv: <https://arxiv.org/abs/1702.01243>\r\n\r\n**Attentional Network for Visual Object Detection**\r\n\r\n- intro: University of Maryland & Mitsubishi Electric Research Laboratories\r\n- arxiv: <https://arxiv.org/abs/1702.01478>\r\n\r\n**Learning Chained Deep Features and Classifiers for Cascade in Object Detection**\r\n\r\n- keykwords: CC-Net\r\n- intro: chained cascade network (CC-Net). 81.1% mAP on PASCAL VOC 2007\r\n- arxiv: <https://arxiv.org/abs/1702.07054>\r\n\r\n**DeNet: Scalable Real-time Object Detection with Directed Sparse Sampling**\r\n\r\n- intro: ICCV 2017 (poster)\r\n- arxiv: <https://arxiv.org/abs/1703.10295>\r\n\r\n**Discriminative Bimodal Networks for Visual Localization and Detection with Natural Language Queries**\r\n\r\n- intro: CVPR 2017\r\n- arxiv: <https://arxiv.org/abs/1704.03944>\r\n\r\n**Spatial Memory for Context Reasoning in Object Detection**\r\n\r\n- arxiv: <https://arxiv.org/abs/1704.04224>\r\n\r\n**Accurate Single Stage Detector Using Recurrent Rolling Convolution**\r\n\r\n- intro: CVPR 2017. SenseTime\r\n- keywords: Recurrent Rolling Convolution (RRC)\r\n- arxiv: <https://arxiv.org/abs/1704.05776>\r\n- github: <https://github.com/xiaohaoChen/rrc_detection>\r\n\r\n**Deep Occlusion Reasoning for Multi-Camera Multi-Target Detection**\r\n\r\n<https://arxiv.org/abs/1704.05775>\r\n\r\n**LCDet: Low-Complexity Fully-Convolutional Neural Networks for Object Detection in Embedded Systems**\r\n\r\n- intro: Embedded Vision Workshop in CVPR. UC San Diego & Qualcomm Inc\r\n- arxiv: <https://arxiv.org/abs/1705.05922>\r\n\r\n**Point Linking Network for Object Detection**\r\n\r\n- intro: Point Linking Network (PLN)\r\n- arxiv: <https://arxiv.org/abs/1706.03646>\r\n\r\n**Perceptual Generative Adversarial Networks for Small Object Detection**\r\n\r\n<https://arxiv.org/abs/1706.05274>\r\n\r\n**Few-shot Object Detection**\r\n\r\n<https://arxiv.org/abs/1706.08249>\r\n\r\n**Yes-Net: An effective Detector Based on Global Information**\r\n\r\n<https://arxiv.org/abs/1706.09180>\r\n\r\n**SMC Faster R-CNN: Toward a scene-specialized multi-object detector**\r\n\r\n<https://arxiv.org/abs/1706.10217>\r\n\r\n**Towards lightweight convolutional neural networks for object detection**\r\n\r\n<https://arxiv.org/abs/1707.01395>\r\n\r\n**RON: Reverse Connection with Objectness Prior Networks for Object Detection**\r\n\r\n- intro: CVPR 2017\r\n- arxiv: <https://arxiv.org/abs/1707.01691>\r\n- github: <https://github.com/taokong/RON>\r\n\r\n**Mimicking Very Efficient Network for Object Detection**\r\n\r\n- intro: CVPR 2017. SenseTime & Beihang University\r\n- paper: <http://openaccess.thecvf.com/content_cvpr_2017/papers/Li_Mimicking_Very_Efficient_CVPR_2017_paper.pdf>\r\n\r\n**Residual Features and Unified Prediction Network for Single Stage Detection**\r\n\r\n<https://arxiv.org/abs/1707.05031>\r\n\r\n**Deformable Part-based Fully Convolutional Network for Object Detection**\r\n\r\n- intro: BMVC 2017 (oral). Sorbonne Universités & CEDRIC\r\n- arxiv: <https://arxiv.org/abs/1707.06175>\r\n\r\n**Adaptive Feeding: Achieving Fast and Accurate Detections by Adaptively Combining Object Detectors**\r\n\r\n- intro: ICCV 2017\r\n- arxiv: <https://arxiv.org/abs/1707.06399>\r\n\r\n**Recurrent Scale Approximation for Object Detection in CNN**\r\n\r\n- intro: ICCV 2017\r\n- keywords: Recurrent Scale Approximation (RSA)\r\n- arxiv: <https://arxiv.org/abs/1707.09531>\r\n- github: <https://github.com/sciencefans/RSA-for-object-detection>\r\n\r\n## DSOD\r\n\r\n**DSOD: Learning Deeply Supervised Object Detectors from Scratch**\r\n\r\n![img](https://user-images.githubusercontent.com/3794909/28934967-718c9302-78b5-11e7-89ee-8b514e53e23c.png)\r\n\r\n- intro: ICCV 2017. Fudan University & Tsinghua University & Intel Labs China\r\n- arxiv: <https://arxiv.org/abs/1708.01241>\r\n- github: <https://github.com/szq0214/DSOD>\r\n- github:https://github.com/Windaway/DSOD-Tensorflow\r\n- github:https://github.com/chenyuntc/dsod.pytorch\r\n\r\n**Learning Object Detectors from Scratch with Gated Recurrent Feature Pyramids**\r\n\r\n- arxiv:https://arxiv.org/abs/1712.00886\r\n- github:https://github.com/szq0214/GRP-DSOD\r\n\r\n**Tiny-DSOD: Lightweight Object Detection for Resource-Restricted Usages**\r\n\r\n- intro: BMVC 2018\r\n- arXiv: https://arxiv.org/abs/1807.11013\r\n\r\n**Object Detection from Scratch with Deep Supervision**\r\n\r\n- intro: This is an extended version of DSOD\r\n- arXiv: https://arxiv.org/abs/1809.09294\r\n\r\n## RetinaNet\r\n\r\n**Focal Loss for Dense Object Detection**\r\n\r\n- intro: ICCV 2017 Best student paper award. Facebook AI Research\r\n- keywords: RetinaNet\r\n- arxiv: <https://arxiv.org/abs/1708.02002>\r\n\r\n**CoupleNet: Coupling Global Structure with Local Parts for Object Detection**\r\n\r\n- intro: ICCV 2017\r\n- arxiv: <https://arxiv.org/abs/1708.02863>\r\n\r\n**Incremental Learning of Object Detectors without Catastrophic Forgetting**\r\n\r\n- intro: ICCV 2017. Inria\r\n- arxiv: <https://arxiv.org/abs/1708.06977>\r\n\r\n**Zoom Out-and-In Network with Map Attention Decision for Region Proposal and Object Detection**\r\n\r\n<https://arxiv.org/abs/1709.04347>\r\n\r\n**StairNet: Top-Down Semantic Aggregation for Accurate One Shot Detection**\r\n\r\n<https://arxiv.org/abs/1709.05788>\r\n\r\n**Dynamic Zoom-in Network for Fast Object Detection in Large Images**\r\n\r\n<https://arxiv.org/abs/1711.05187>\r\n\r\n**Zero-Annotation Object Detection with Web Knowledge Transfer**\r\n\r\n- intro: NTU, Singapore & Amazon\r\n- keywords: multi-instance multi-label domain adaption learning framework\r\n- arxiv: <https://arxiv.org/abs/1711.05954>\r\n\r\n## MegDet\r\n\r\n**MegDet: A Large Mini-Batch Object Detector**\r\n\r\n- intro: Peking University & Tsinghua University & Megvii Inc\r\n- arxiv: <https://arxiv.org/abs/1711.07240>\r\n\r\n**Receptive Field Block Net for Accurate and Fast Object Detection**\r\n\r\n- intro: RFBNet\r\n- arxiv: <https://arxiv.org/abs/1711.07767>\r\n- github: <https://github.com//ruinmessi/RFBNet>\r\n\r\n**An Analysis of Scale Invariance in Object Detection - SNIP**\r\n\r\n- arxiv: <https://arxiv.org/abs/1711.08189>\r\n- github: <https://github.com/bharatsingh430/snip>\r\n\r\n**Feature Selective Networks for Object Detection**\r\n\r\n<https://arxiv.org/abs/1711.08879>\r\n\r\n**Learning a Rotation Invariant Detector with Rotatable Bounding Box**\r\n\r\n- arxiv: <https://arxiv.org/abs/1711.09405>\r\n- github: <https://github.com/liulei01/DRBox>\r\n\r\n**Scalable Object Detection for Stylized Objects**\r\n\r\n- intro: Microsoft AI & Research Munich\r\n- arxiv: <https://arxiv.org/abs/1711.09822>\r\n\r\n**Learning Object Detectors from Scratch with Gated Recurrent Feature Pyramids**\r\n\r\n- arxiv: <https://arxiv.org/abs/1712.00886>\r\n- github: <https://github.com/szq0214/GRP-DSOD>\r\n\r\n**Deep Regionlets for Object Detection**\r\n\r\n- keywords: region selection network, gating network\r\n- arxiv: <https://arxiv.org/abs/1712.02408>\r\n\r\n**Training and Testing Object Detectors with Virtual Images**\r\n\r\n- intro: IEEE/CAA Journal of Automatica Sinica\r\n- arxiv: <https://arxiv.org/abs/1712.08470>\r\n\r\n**Large-Scale Object Discovery and Detector Adaptation from Unlabeled Video**\r\n\r\n- keywords: object mining, object tracking, unsupervised object discovery by appearance-based clustering, self-supervised detector adaptation\r\n- arxiv: <https://arxiv.org/abs/1712.08832>\r\n\r\n**Spot the Difference by Object Detection**\r\n\r\n- intro: Tsinghua University & JD Group\r\n- arxiv: <https://arxiv.org/abs/1801.01051>\r\n\r\n**Localization-Aware Active Learning for Object Detection**\r\n\r\n- arxiv: <https://arxiv.org/abs/1801.05124>\r\n\r\n**Object Detection with Mask-based Feature Encoding**\r\n\r\n- arxiv: <https://arxiv.org/abs/1802.03934>\r\n\r\n**LSTD: A Low-Shot Transfer Detector for Object Detection**\r\n\r\n- intro: AAAI 2018\r\n- arxiv: <https://arxiv.org/abs/1803.01529>\r\n\r\n**Pseudo Mask Augmented Object Detection**\r\n\r\n<https://arxiv.org/abs/1803.05858>\r\n\r\n**Revisiting RCNN: On Awakening the Classification Power of Faster RCNN**\r\n\r\n<https://arxiv.org/abs/1803.06799>\r\n\r\n**Learning Region Features for Object Detection**\r\n\r\n- intro: Peking University & MSRA\r\n- arxiv: <https://arxiv.org/abs/1803.07066>\r\n\r\n**Single-Shot Bidirectional Pyramid Networks for High-Quality Object Detection**\r\n\r\n- intro: Singapore Management University & Zhejiang University\r\n- arxiv: <https://arxiv.org/abs/1803.08208>\r\n\r\n**Object Detection for Comics using Manga109 Annotations**\r\n\r\n- intro: University of Tokyo & National Institute of Informatics, Japan\r\n- arxiv: <https://arxiv.org/abs/1803.08670>\r\n\r\n**Task-Driven Super Resolution: Object Detection in Low-resolution Images**\r\n\r\n- arxiv: <https://arxiv.org/abs/1803.11316>\r\n\r\n**Transferring Common-Sense Knowledge for Object Detection**\r\n\r\n- arxiv: <https://arxiv.org/abs/1804.01077>\r\n\r\n**Multi-scale Location-aware Kernel Representation for Object Detection**\r\n\r\n- intro: CVPR 2018\r\n- arxiv: <https://arxiv.org/abs/1804.00428>\r\n- github: <https://github.com/Hwang64/MLKP>\r\n\r\n\r\n**Loss Rank Mining: A General Hard Example Mining Method for Real-time Detectors**\r\n\r\n- intro: National University of Defense Technology\r\n- arxiv: https://arxiv.org/abs/1804.04606\r\n\r\n**Robust Physical Adversarial Attack on Faster R-CNN Object Detector**\r\n\r\n- arxiv: https://arxiv.org/abs/1804.05810\r\n\r\n## RefineNet\r\n\r\n**Single-Shot Refinement Neural Network for Object Detection**\r\n\r\n- intro: CVPR 2018\r\n\r\n- arxiv: <https://arxiv.org/abs/1711.06897>\r\n- github: <https://github.com/sfzhang15/RefineDet>\r\n- github: https://github.com/lzx1413/PytorchSSD\r\n- github: https://github.com/ddlee96/RefineDet_mxnet\r\n- github: https://github.com/MTCloudVision/RefineDet-Mxnet\r\n\r\n## DetNet\r\n\r\n**DetNet: A Backbone network for Object Detection**\r\n\r\n- intro: Tsinghua University & Face++\r\n- arxiv: https://arxiv.org/abs/1804.06215\r\n\r\n\r\n## SSOD\r\n\r\n**Self-supervisory Signals for Object Discovery and Detection**\r\n\r\n- Google Brain\r\n- arxiv:https://arxiv.org/abs/1806.03370\r\n\r\n## CornerNet\r\n\r\n**CornerNet: Detecting Objects as Paired Keypoints**\r\n\r\n- intro: ECCV 2018\r\n- arXiv: https://arxiv.org/abs/1808.01244\r\n- github: <https://github.com/umich-vl/CornerNet>\r\n\r\n## M2Det\r\n\r\n**M2Det: A Single-Shot Object Detector based on Multi-Level Feature Pyramid Network**\r\n\r\n- intro: AAAI 2019\r\n- arXiv: https://arxiv.org/abs/1811.04533\r\n- github: https://github.com/qijiezhao/M2Det\r\n\r\n## 3D Object Detection\r\n\r\n**3D Backbone Network for 3D Object Detection**\r\n\r\n- arXiv: https://arxiv.org/abs/1901.08373\r\n\r\n**LMNet: Real-time Multiclass Object Detection on CPU using 3D LiDARs**\r\n\r\n- arxiv: https://arxiv.org/abs/1805.04902\r\n- github: https://github.com/CPFL/Autoware/tree/feature/cnn_lidar_detection\r\n\r\n\r\n## ZSD（Zero-Shot Object Detection）\r\n\r\n**Zero-Shot Detection**\r\n\r\n- intro: Australian National University\r\n- keywords: YOLO\r\n- arxiv: <https://arxiv.org/abs/1803.07113>\r\n\r\n**Zero-Shot Object Detection**\r\n\r\n- arxiv: https://arxiv.org/abs/1804.04340\r\n\r\n**Zero-Shot Object Detection: Learning to Simultaneously Recognize and Localize Novel Concepts**\r\n\r\n- arxiv: https://arxiv.org/abs/1803.06049\r\n\r\n**Zero-Shot Object Detection by Hybrid Region Embedding**\r\n\r\n- arxiv: https://arxiv.org/abs/1805.06157\r\n\r\n## OSD（One-Shot Object Detection）\r\n\r\n**Comparison Network for One-Shot Conditional Object Detection**\r\n\r\n- arXiv: https://arxiv.org/abs/1904.02317\r\n\r\n**One-Shot Object Detection**\r\n\r\nRepMet: Representative-based metric learning for classification and one-shot object detection\r\n\r\n- intro: IBM Research AI\r\n- arxiv:https://arxiv.org/abs/1806.04728\r\n- github: TODO\r\n\r\n## Weakly Supervised Object Detection\r\n\r\n**Weakly Supervised Object Detection in Artworks**\r\n\r\n- intro: ECCV 2018 Workshop Computer Vision for Art Analysis\r\n- arXiv: https://arxiv.org/abs/1810.02569\r\n- Datasets: https://wsoda.telecom-paristech.fr/downloads/dataset/IconArt_v1.zip\r\n\r\n**Cross-Domain Weakly-Supervised Object Detection through Progressive Domain Adaptation**\r\n\r\n- intro: CVPR 2018\r\n- arXiv: https://arxiv.org/abs/1803.11365\r\n- homepage: https://naoto0804.github.io/cross_domain_detection/\r\n- paper: http://openaccess.thecvf.com/content_cvpr_2018/html/Inoue_Cross-Domain_Weakly-Supervised_Object_CVPR_2018_paper.html\r\n- github: https://github.com/naoto0804/cross-domain-detection\r\n\r\n## Softer-NMS\r\n\r\n**《Softer-NMS: Rethinking Bounding Box Regression for Accurate Object Detection》**\r\n\r\n- intro: CMU & Face++\r\n- arXiv: https://arxiv.org/abs/1809.08545\r\n- github: https://github.com/yihui-he/softer-NMS\r\n\r\n## 2019\r\n\r\n**Feature Selective Anchor-Free Module for Single-Shot Object Detection**\r\n\r\n- intro: CVPR 2019\r\n\r\n- arXiv: https://arxiv.org/abs/1903.00621\r\n\r\n**Object Detection based on Region Decomposition and Assembly**\r\n\r\n- intro: AAAI 2019\r\n\r\n- arXiv: https://arxiv.org/abs/1901.08225\r\n\r\n**Bottom-up Object Detection by Grouping Extreme and Center Points**\r\n\r\n- intro: one stage 43.2% on COCO test-dev\r\n- arXiv: https://arxiv.org/abs/1901.08043\r\n- github: https://github.com/xingyizhou/ExtremeNet\r\n\r\n**ORSIm Detector: A Novel Object Detection Framework in Optical Remote Sensing Imagery Using Spatial-Frequency Channel Features**\r\n\r\n- intro: IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING\r\n\r\n- arXiv: https://arxiv.org/abs/1901.07925\r\n\r\n**Consistent Optimization for Single-Shot Object Detection**\r\n\r\n- intro: improves RetinaNet from 39.1 AP to 40.1 AP on COCO datase\r\n\r\n- arXiv: https://arxiv.org/abs/1901.06563\r\n\r\n**Learning Pairwise Relationship for Multi-object Detection in Crowded Scenes**\r\n\r\n- arXiv: https://arxiv.org/abs/1901.03796\r\n\r\n**RetinaMask: Learning to predict masks improves state-of-the-art single-shot detection for free**\r\n\r\n- arXiv: https://arxiv.org/abs/1901.03353\r\n- github: https://github.com/chengyangfu/retinamask\r\n\r\n**Region Proposal by Guided Anchoring**\r\n\r\n- intro: CUHK - SenseTime Joint Lab\r\n- arXiv: https://arxiv.org/abs/1901.03278\r\n\r\n**Scale-Aware Trident Networks for Object Detection**\r\n\r\n- intro: mAP of **48.4** on the COCO dataset\r\n- arXiv: https://arxiv.org/abs/1901.01892\r\n\r\n## 2018\r\n\r\n**Large-Scale Object Detection of Images from Network Cameras in Variable Ambient Lighting Conditions**\r\n\r\n- arXiv: https://arxiv.org/abs/1812.11901\r\n\r\n**Strong-Weak Distribution Alignment for Adaptive Object Detection**\r\n\r\n- arXiv: https://arxiv.org/abs/1812.04798\r\n\r\n**AutoFocus: Efficient Multi-Scale Inference**\r\n\r\n- intro: AutoFocus obtains an **mAP of 47.9%** (68.3% at 50% overlap) on the **COCO test-dev** set while processing **6.4 images per second on a Titan X (Pascal) GPU** \r\n- arXiv: https://arxiv.org/abs/1812.01600\r\n\r\n**NOTE-RCNN: NOise Tolerant Ensemble RCNN for Semi-Supervised Object Detection**\r\n\r\n- intro: Google Could\r\n- arXiv: https://arxiv.org/abs/1812.00124\r\n\r\n**SPLAT: Semantic Pixel-Level Adaptation Transforms for Detection**\r\n\r\n- intro: UC Berkeley\r\n- arXiv: https://arxiv.org/abs/1812.00929\r\n\r\n**Grid R-CNN**\r\n\r\n- intro: SenseTime\r\n- arXiv: https://arxiv.org/abs/1811.12030\r\n\r\n**Deformable ConvNets v2: More Deformable, Better Results**\r\n\r\n- intro: Microsoft Research Asia\r\n\r\n- arXiv: https://arxiv.org/abs/1811.11168\r\n\r\n**Anchor Box Optimization for Object Detection**\r\n\r\n- intro: Microsoft Research\r\n- arXiv: https://arxiv.org/abs/1812.00469\r\n\r\n**Efficient Coarse-to-Fine Non-Local Module for the Detection of Small Objects**\r\n\r\n- intro: https://arxiv.org/abs/1811.12152\r\n\r\n**NOTE-RCNN: NOise Tolerant Ensemble RCNN for Semi-Supervised Object Detection**\r\n\r\n- arXiv: https://arxiv.org/abs/1812.00124\r\n\r\n**Learning RoI Transformer for Detecting Oriented Objects in Aerial Images**\r\n\r\n- arXiv: https://arxiv.org/abs/1812.00155\r\n\r\n**Integrated Object Detection and Tracking with Tracklet-Conditioned Detection**\r\n\r\n- intro: Microsoft Research Asia\r\n- arXiv: https://arxiv.org/abs/1811.11167\r\n\r\n**Deep Regionlets: Blended Representation and Deep Learning for Generic Object Detection**\r\n\r\n- arXiv: https://arxiv.org/abs/1811.11318\r\n\r\n **Gradient Harmonized Single-stage Detector**\r\n\r\n- intro: AAAI 2019\r\n- arXiv: https://arxiv.org/abs/1811.05181\r\n\r\n**CFENet: Object Detection with Comprehensive Feature Enhancement Module**\r\n\r\n- intro: ACCV 2018\r\n- github: https://github.com/qijiezhao/CFENet\r\n\r\n**DeRPN: Taking a further step toward more general object detection**\r\n\r\n- intro: AAAI 2019\r\n- arXiv: https://arxiv.org/abs/1811.06700\r\n- github: https://github.com/HCIILAB/DeRPN\r\n\r\n**Hybrid Knowledge Routed Modules for Large-scale Object Detection**\r\n\r\n- intro: Sun Yat-Sen University & Huawei Noah’s Ark Lab\r\n- arXiv: https://arxiv.org/abs/1810.12681\r\n- github: https://github.com/chanyn/HKRM\r\n\r\n**《Receptive Field Block Net for Accurate and Fast Object Detection》**\r\n\r\n- intro: ECCV 2018\r\n- arXiv: [https://arxiv.org/abs/1711.07767](https://arxiv.org/abs/1711.07767)\r\n- github: [https://github.com/ruinmessi/RFBNet](https://github.com/ruinmessi/RFBNet)\r\n\r\n**Deep Feature Pyramid Reconfiguration for Object Detection**\r\n\r\n- intro: ECCV 2018\r\n- arXiv: https://arxiv.org/abs/1808.07993\r\n\r\n**Unsupervised Hard Example Mining from Videos for Improved Object Detection**\r\n\r\n- intro: ECCV 2018\r\n- arXiv: https://arxiv.org/abs/1808.04285\r\n\r\n**Acquisition of Localization Confidence for Accurate Object Detection**\r\n\r\n- intro: ECCV 2018\r\n- arXiv: https://arxiv.org/abs/1807.11590\r\n- github: https://github.com/vacancy/PreciseRoIPooling\r\n\r\n**Toward Scale-Invariance and Position-Sensitive Region Proposal Networks**\r\n\r\n- intro: ECCV 2018\r\n- arXiv: https://arxiv.org/abs/1807.09528\r\n\r\n**MetaAnchor: Learning to Detect Objects with Customized Anchors**\r\n\r\n- arxiv: https://arxiv.org/abs/1807.00980\r\n\r\n**Relation Network for Object Detection**\r\n\r\n- intro: CVPR 2018\r\n- arxiv: https://arxiv.org/abs/1711.11575\r\n- github:https://github.com/msracver/Relation-Networks-for-Object-Detection\r\n\r\n**Quantization Mimic: Towards Very Tiny CNN for Object Detection**\r\n\r\n- Tsinghua University1 & The Chinese University of Hong Kong2 &SenseTime3\r\n- arxiv: https://arxiv.org/abs/1805.02152\r\n\r\n**Learning Rich Features for Image Manipulation Detection**\r\n\r\n- intro: CVPR 2018 Camera Ready\r\n- arxiv: https://arxiv.org/abs/1805.04953\r\n\r\n**SNIPER: Efficient Multi-Scale Training**\r\n\r\n- arxiv:https://arxiv.org/abs/1805.09300\r\n- github:https://github.com/mahyarnajibi/SNIPER\r\n\r\n**Soft Sampling for Robust Object Detection**\r\n\r\n- intro: the robustness of object detection under the presence of missing annotations\r\n- arxiv:https://arxiv.org/abs/1806.06986\r\n\r\n**Cost-effective Object Detection: Active Sample Mining with Switchable Selection Criteria**\r\n\r\n- intro: TNNLS 2018\r\n- arxiv:https://arxiv.org/abs/1807.00147\r\n- code: http://kezewang.com/codes/ASM_ver1.zip\r\n\r\n## Other\r\n\r\n**R3-Net: A Deep Network for Multi-oriented Vehicle Detection in Aerial Images and Videos**\r\n\r\n- arxiv: https://arxiv.org/abs/1808.05560\r\n- youtube: https://youtu.be/xCYD-tYudN0\r\n\r\n# Detection Toolbox\r\n\r\n- [Detectron(FAIR)](https://github.com/facebookresearch/Detectron): Detectron is Facebook AI Research's software system that implements state-of-the-art object detection algorithms, including [Mask R-CNN](https://arxiv.org/abs/1703.06870). It is written in Python and powered by the [Caffe2](https://github.com/caffe2/caffe2) deep learning framework.\r\n- [Detectron2](https://github.com/facebookresearch/detectron2): Detectron2 is FAIR's next-generation research platform for object detection and segmentation.\r\n- [maskrcnn-benchmark(FAIR)](https://github.com/facebookresearch/maskrcnn-benchmark): Fast, modular reference implementation of Instance Segmentation and Object Detection algorithms in PyTorch.\r\n- [mmdetection(SenseTime&CUHK)](https://github.com/open-mmlab/mmdetection): mmdetection is an open source object detection toolbox based on PyTorch. It is a part of the open-mmlab project developed by [Multimedia Laboratory, CUHK](http://mmlab.ie.cuhk.edu.hk/).\r\n"
  }
]