本次分享的是所有CVPR2020檢測類論文，并將它們細(xì)分為3D目標(biāo)檢測、人臉檢測、動作檢測、視頻目標(biāo)檢測、文本檢測、行人檢測等方向。

3D目標(biāo)檢測

【1】Learning Deep Network for Detecting 3D Object Keypoints and 6D Poses

作者：Wanqing Zhao, Shaobo Zhang, Ziyu Guan, Wei Zhao, Jinye Peng, Jianping Fan

【2】DOPS: Learning to Detect 3D Objects and Predict Their 3D Shapes

作者：Mahyar Najibi, Guangda Lai, Abhijit Kundu, Zhichao Lu, Vivek Rathod, Thomas Funkhouser, Caroline Pantofaru, David Ross, Larry S. Davis, Alireza Fathi

【3】Train in Germany, Test in the USA: Making 3D Object Detectors Generalize

作者：Yan Wang, Xiangyu Chen, Yurong You, Li Erran Li, Bharath Hariharan, Mark Campbell, Kilian Q. Weinberger, Wei-Lun Chao

代碼：https://github.com/cxy1997/3D_adapt_auto_driving

【4】3DSSD: Point-Based 3D Single Stage Object Detector

作者：Zetong Yang, Yanan Sun, Shu Liu, Jiaya Jia

代碼：https://github.com/tomztyang/3DSSD

本文主要從point-based的研究入手，考慮如何解決掉以前的point-based的方法的瓶頸，即時間和內(nèi)存占有遠(yuǎn)遠(yuǎn)大于voxel-based的方法，從而作者設(shè)計了新的SA模塊和丟棄了FP模塊到達時間上可達25FPS，此外本文采用一個anchor free Head，進一步減少時間和GPU顯存，提出了3D center-ness label的表示，進一步提高了精度。

【5】FroDO: From Detections to 3D Objects

作者：Martin Runz, Kejie Li, Meng Tang, Lingni Ma, Chen Kong, Tanner Schmidt, Ian Reid, Lourdes Agapito, Julian Straub, Steven Lovegrove, Richard Newcombe

【6】Associate-3Ddet: Perceptual-to-Conceptual Association for 3D Point Cloud Object Detection

作者：Liang Du, Xiaoqing Ye, Xiao Tan, Jianfeng Feng, Zhenbo Xu, Errui Ding, Shilei Wen

【7】IDA-3D: Instance-Depth-Aware 3D Object Detection From Stereo Vision for Autonomous Driving

作者：Wanli Peng, Hao Pan, He Liu, Yi Sun

【8】DSGN: Deep Stereo Geometry Network for 3D Object Detection

作者：Yilun Chen, Shu Liu, Xiaoyong Shen, Jiaya Jia

代碼：https://github.com/chenyilun95/DSGN

【9】DR Loss: Improving Object Detection by Distributional Ranking

作者：Qi Qian, Lei Chen, Hao Li, Rong Jin

【10】MonoPair: Monocular 3D Object Detection Using Pairwise Spatial Relationships

作者：Yongjian Chen, Lei Tai, Kai Sun, Mingyang Li

【11】Structure Aware Single-Stage 3D Object Detection From Point Cloud

作者：Chenhang He, Hui Zeng, Jianqiang Huang, Xian-Sheng Hua, Lei Zhang

【12】Learning Depth-Guided Convolutions for Monocular 3D Object Detection

作者：Mingyu Ding, Yuqi Huo, Hongwei Yi, Zhe Wang, Jianping Shi, Zhiwu Lu, Ping Luo

【13】LiDAR-Based Online 3D Video Object Detection With Graph-Based Message Passing and Spatiotemporal Transformer Attention

作者：Junbo Yin, Jianbing Shen, Chenye Guan, Dingfu Zhou, Ruigang Yang

【14】SESS: Self-Ensembling Semi-Supervised 3D Object Detection作者：Na Zhao, Tat-Seng Chua, Gim Hee Lee

【15】What You See is What You Get: Exploiting Visibility for 3D Object Detection

作者：Peiyun Hu, Jason Ziglar, David Held, Deva Ramanan

【16】Density-Based Clustering for 3D Object Detection in Point Clouds

作者：Syeda Mariam Ahmed, Chee Meng Chew

【17】Disp R-CNN: Stereo 3D Object Detection via Shape Prior Guided Instance Disparity Estimation

作者：Jiaming Sun, Linghao Chen, Yiming Xie, Siyu Zhang, Qinhong Jiang, Xiaowei Zhou, Hujun Bao

代碼：https://github.com/zju3dv/disprcn

【18】PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detection

作者：Shaoshuai Shi, Chaoxu Guo, Li Jiang, Zhe Wang, Jianping Shi, Xiaogang Wang, Hongsheng Li

代碼：https://github.com/sshaoshuai/PV-RCNN

【19】MLCVNet: Multi-Level Context VoteNet for 3D Object Detection

作者：Qian Xie, Yu-Kun Lai, Jing Wu, Zhoutao Wang, Yiming Zhang, Kai Xu, Jun Wang

代碼：https://github.com/NUAAXQ/MLCVNet

【20】A Hierarchical Graph Network for 3D Object Detection on Point Clouds

作者：Jintai Chen, Biwen Lei, Qingyu Song, Haochao Ying, Danny Z. Chen, Jian Wu

【21】HVNet: Hybrid Voxel Network for LiDAR Based 3D Object Detection

作者：Maosheng Ye, Shuangjie Xu, Tongyi Cao

3D目標(biāo)檢測是當(dāng)前自動駕駛感知模塊重要的一個環(huán)節(jié)，如何平衡3D物體檢測的精度以及速度更是非常重要的一個研究話題。本文提出了一種新的基于點云的三維物體檢測的統(tǒng)一網(wǎng)絡(luò)：混合體素網(wǎng)絡(luò)(HVNet)，通過在點級別上混合尺度體素特征編碼器(VFE)得到更好的體素特征編碼方法，從而在速度和精度上得到提升。與多種方法相比，HVNet在檢測速度上有明顯的提高。在KITTI 數(shù)據(jù)集自行車檢測的中等難度級別（moderate）中，HVNet 的準(zhǔn)確率比PointPillars方法高出了8.44%。

【22】Point-GNN: Graph Neural Network for 3D Object Detection in a Point Cloud

作者：Weijing Shi, Raj Rajkumar

代碼：https://github.com/WeijingShi/Point-GNN

【23】Joint 3D Instance Segmentation and Object Detection for Autonomous Driving

作者：Dingfu Zhou, Jin Fang, Xibin Song, Liu Liu, Junbo Yin, Yuchao Dai, Hongdong Li, Ruigang Yang

【24】FocalMix: Semi-Supervised Learning for 3D Medical Image Detection

作者：Dong Wang, Yuan Zhang, Kexin Zhang, Liwei Wang

【25】ImVoteNet: Boosting 3D Object Detection in Point Clouds With Image Votes作者：Charles R. Qi, Xinlei Chen, Or Litany, Leonidas J. Guibas

【26】PointPainting: Sequential Fusion for 3D Object Detection

作者：Sourabh Vora, Alex H. Lang, Bassam Helou, Oscar Beijbom

【27】End-to-End Pseudo-LiDAR for Image-Based 3D Object Detection

作者：Rui Qian, Divyansh Garg, Yan Wang, Yurong You, Serge Belongie, Bharath Hariharan, Mark Campbell, Kilian Q. Weinberger, Wei-Lun Chao

代碼：https://github.com/mileyan/pseudo-LiDAR_e2e

人物（交互）檢測

【28】Learning Human-Object Interaction Detection Using Interaction Points

作者：Tiancai Wang, Tong Yang, Martin Danelljan, Fahad Shahbaz Khan, Xiangyu Zhang, Jian Sun

【29】PPDM: Parallel Point Detection and Matching for Real-Time Human-Object Interaction Detection

作者：Yue Liao, Si Liu, Fei Wang, Yanjie Chen, Chen Qian, Jiashi Feng

代碼：https://github.com/YueLiao/PPDM

【30】（人物檢測）Learning to Detect Important People in Unlabelled Images for Semi-Supervised Important People Detection

作者：Fa-Ting Hong, Wei-Hong Li, Wei-Shi Zheng

【31】（人體檢測）VSGNet: Spatial Attention Network for Detecting Human Object Interactions Using Graph Convolutions

作者：Oytun Ulutan, A S M Iftekhar, B. S. Manjunath

動作檢測

【32】Combining Detection and Tracking for Human Pose Estimation in Videos

作者：Manchen Wang, Joseph Tighe, Davide Modolo

【33】G-TAD: Sub-Graph Localization for Temporal Action Detection

作者：Mengmeng Xu, Chen Zhao, David S. Rojas, Ali Thabet, Bernard Ghanem

【34】Learning to Discriminate Information for Online Action Detection

作者：Hyunjun Eun, Jinyoung Moon, Jongyoul Park, Chanho Jung, Changick Kim

活體檢測

【35】ZSTAD: Zero-Shot Temporal Activity Detection

作者：Lingling Zhang, Xiaojun Chang, Jun Liu, Minnan Luo, Sen Wang, Zongyuan Ge, Alexander Hauptmann

顯著性檢測

【36】Learning Selective Self-Mutual Attention for RGB-D Saliency Detection

作者：Nian Liu, Ni Zhang, Junwei Han

【37】Label Decoupling Framework for Salient Object Detection

作者：Jun Wei, Shuhui Wang, Zhe Wu, Chi Su, Qingming Huang, Qi Tian

【38】Weakly-Supervised Salient Object Detection via Scribble Annotations

作者：Jing Zhang, Xin Yu, Aixuan Li, Peipei Song, Bowen Liu, Yuchao Dai

【39】UC-Net: Uncertainty Inspired RGB-D Saliency Detection via Conditional Variational Autoencoders

作者：Jing Zhang, Deng-Ping Fan, Yuchao Dai, Saeed Anwar, Fatemeh Sadat Saleh, Tong Zhang, Nick Barnes

代碼：https://github.com/JingZhang617/UCNet

【40】Adaptive Graph Convolutional Network With Attention Graph Clustering for Co-Saliency Detection

作者：Kaihua Zhang, Tengpeng Li, Shiwen Shen, Bo Liu, Jin Chen, Qingshan Liu

【41】A2dele: Adaptive and Attentive Depth Distiller for Efficient RGB-D Salient Object Detection

作者：Yongri Piao, Zhengkun Rong, Miao Zhang, Weisong Ren, Huchuan Lu

【42】Interactive Two-Stream Decoder for Accurate and Fast Saliency Detection

作者：Huajun Zhou, Xiaohua Xie, Jian-Huang Lai, Zixuan Chen, Lingxiao Yang

【43】Multi-Scale Interactive Network for Salient Object Detection

作者：Youwei Pang, Xiaoqi Zhao, Lihe Zhang, Huchuan Lu

【44】Taking a Deeper Look at Co-Salient Object Detection

作者：Deng-Ping Fan, Zheng Lin, Ge-Peng Ji, Dingwen Zhang, Huazhu Fu, Ming-Ming Cheng

【45】JL-DCF: Joint Learning and Densely-Cooperative Fusion Framework for RGB-D Salient Object Detection

作者：Keren Fu, Deng-Ping Fan, Ge-Peng Ji, Qijun Zhao

代碼：https://github.com/kerenfu/JLDCF/

【46】Select, Supplement and Focus for RGB-D Saliency Detection

作者：Miao Zhang, Weisong Ren, Yongri Piao, Zhengkun Rong, Huchuan Lu

偽裝/偽造檢測

【47】Camouflaged Object Detection

作者：Deng-Ping Fan, Ge-Peng Ji, Guolei Sun, Ming-Ming Cheng, Jianbing Shen, Ling Shao

【48】DOA-GAN: Dual-Order Attentive Generative Adversarial Network for Image Copy-Move Forgery Detection and Localization

作者：Ashraful Islam, Chengjiang Long, Arslan Basharat, Anthony Hoogs

【49】Advancing High Fidelity Identity Swapping for Forgery Detection

作者：Lingzhi Li, Jianmin Bao, Hao Yang, Dong Chen, Fang Wen

【50】Advancing High Fidelity Identity Swapping for Forgery Detection

作者：Lingzhi Li, Jianmin Bao, Hao Yang, Dong Chen, Fang Wen

人臉檢測

【51】Cross-Domain Face Presentation Attack Detection via Multi-Domain Disentangled Representation Learning

作者：Guoqing Wang, Hu Han, Shiguang Shan, Xilin Chen

【52】HAMBox: Delving Into Mining High-Quality Anchors on Face Detection

作者：Yang Liu, Xu Tang, Junyu Han, Jingtuo Liu, Dinger Rui, Xiang Wu

【53】BFBox: Searching Face-Appropriate Backbone and Feature Pyramid Network for Face Detector

作者：Yang Liu, Xu Tang

【54】Global Texture Enhancement for Fake Face Detection in the Wild

作者：Zhengzhe Liu, Xiaojuan Qi, Philip H.S. Torr

【55】（數(shù)據(jù)集）DeeperForensics-1.0: A Large-Scale Dataset for Real-World Face Forgery Detection

作者：Liming Jiang, Ren Li, Wayne Wu, Chen Qian, Chen Change Loy

【56】Face X-Ray for More General Face Forgery Detection

作者：Lingzhi Li, Jianmin Bao, Ting Zhang, Hao Yang, Dong Chen, Fang Wen, Baining Guo

【57】On the Detection of Digital Face Manipulation

作者：Hao Dang, Feng Liu, Joel Stehouwer, Xiaoming Liu, Anil K. Jain

【58】Attention-Driven Cropping for Very High Resolution Facial Landmark Detection

作者：Prashanth Chandran, Derek Bradley, Markus Gross, Thabo Beeler

小樣本/零樣本

【59】Few-Shot Object Detection With Attention-RPN and Multi-Relation Detector

作者：Qi Fan, Wei Zhuo, Chi-Keung Tang, Yu-Wing Tai

本文提出了新的少樣本目標(biāo)檢測算法，創(chuàng)新點包括Attention-RPN、多關(guān)系檢測器以及對比訓(xùn)練策略，另外還構(gòu)建了包含1000類的少樣本檢測數(shù)據(jù)集FSOD，在FSOD上訓(xùn)練得到的論文模型能夠直接遷移到新類別的檢測中，不需要fine-tune。

【60】Incremental Few-Shot Object Detection

作者：Juan-Manuel Perez-Rua, Xiatian Zhu, Timothy M. Hospedales, Tao Xiang

【61】Don't Even Look Once: Synthesizing Features for Zero-Shot Detection

作者：Pengkai Zhu, Hanxiao Wang, Venkatesh Saligrama

異常檢測

【62】Uninformed Students: Student-Teacher Anomaly Detection With Discriminative Latent Embeddings

作者：Paul Bergmann, Michael Fauser, David Sattlegger, Carsten Steger

【63】Graph Embedded Pose Clustering for Anomaly Detection

作者：Amir Markovitz, Gilad Sharir, Itamar Friedman, Lihi Zelnik-Manor, Shai Avidan

【64】Self-Trained Deep Ordinal Regression for End-to-End Video Anomaly Detection

作者：Guansong Pang, Cheng Yan, Chunhua Shen, Anton van den Hengel, Xiao Bai

【65】Learning Memory-Guided Normality for Anomaly Detection

作者：Hyunjong Park, Jongyoun Noh, Bumsub Ham

半監(jiān)督/弱監(jiān)督/無監(jiān)督

【66】DUNIT: Detection-Based Unsupervised Image-to-Image Translation

作者：Deblina Bhattacharjee, Seungryong Kim, Guillaume Vizier, Mathieu Salzmann

【67】A Multi-Task Mean Teacher for Semi-Supervised Shadow Detection

作者：Zhihao Chen, Lei Zhu, Liang Wan, Song Wang, Wei Feng, Pheng-Ann Heng

【68】Instance-Aware, Context-Focused, and Memory-Efficient Weakly Supervised Object Detection

作者：Zhongzheng Ren, Zhiding Yu, Xiaodong Yang, Ming-Yu Liu, Yong Jae Lee, Alexander G. Schwing, Jan Kautz

代碼：https://github.com/NVlabs/wetectron

【69】SLV: Spatial Likelihood Voting for Weakly Supervised Object Detection

作者：Ze Chen, Zhihang Fu, Rongxin Jiang, Yaowu Chen, Xian-Sheng Hua

密集檢測

【70】D3Feat: Joint Learning of Dense Detection and Description of 3D Local Features

作者：Xuyang Bai, Zixin Luo, Lei Zhou, Hongbo Fu, Long Quan, Chiew-Lan Tai

【71】Real-Time Panoptic Segmentation From Dense Detections

作者：Rui Hou, Jie Li, Arjun Bhargava, Allan Raventos, Vitor Guizilini, Chao Fang, Jerome Lynch, Adrien Gaidon

文本檢測

【72】Deep Relational Reasoning Graph Network for Arbitrary Shape Text Detection

作者：Shi-Xue Zhang, Xiaobin Zhu, Jie-Bo Hou, Chang Liu, Chun Yang, Hongfa Wang, Xu-Cheng Yin

【73】ContourNet: Taking a Further Step Toward Accurate Arbitrary-Shaped Scene Text Detection

作者：Yuxin Wang, Hongtao Xie, Zheng-Jun Zha, Mengting Xing, Zilong Fu, Yongdong Zhang

視頻目標(biāo)檢測

【74】Memory Enhanced Global-Local Aggregation for Video Object Detection

作者：Yihong Chen, Yue Cao, Han Hu, Liwei Wang

【75】Beyond Short-Term Snippet: Video Relation Detection With Spatio-Temporal Global Context

作者：Chenchen Liu, Yang Jin, Kehan Xu, Guoqiang Gong, Yadong Mu

【76】Detecting Attended Visual Targets in Video

作者：Eunji Chong, Yongxin Wang, Nataniel Ruiz, James M. Rehg

【77】LiDAR-Based Online 3D Video Object Detection With Graph-Based Message Passing and Spatiotemporal Transformer Attention

作者：Junbo Yin, Jianbing Shen, Chenye Guan, Dingfu Zhou, Ruigang Yang

代碼：https://github.com/yinjunbo/3DVID

【78】Combining Detection and Tracking for Human Pose Estimation in Videos

作者：Manchen Wang, Joseph Tighe, Davide Modolo

行人檢測

【79】STINet: Spatio-Temporal-Interactive Network for Pedestrian Detection and Trajectory Prediction

作者：Zhishuai Zhang, Jiyang Gao, Junhua Mao, Yukai Liu, Dragomir Anguelov, Congcong Li

【80】Temporal-Context Enhanced Detection of Heavily Occluded Pedestrians

作者：Jialian Wu, Chunluan Zhou, Ming Yang, Qian Zhang, yuan Li, Junsong Yuan

【81】Where, What, Whether: Multi-Modal Learning Meets Pedestrian Detection

作者：Yan Luo, Chongyang Zhang, Muming Zhao, Hao Zhou, Jun Sun

【82】NMS by Representative Region: Towards Crowded Pedestrian Detection by Proposal Pairing

作者：Xin Huang, Zheng Ge, Zequn Jie, Osamu Yoshie

移動目標(biāo)檢測

【83】MnasFPN: Learning Latency-Aware Pyramid Architecture for Object Detection on Mobile Devices

作者：Bo Chen, Golnaz Ghiasi, Hanxiao Liu, Tsung-Yi Lin, Dmitry Kalenichenko, Hartwig Adam, Quoc V. Le

通用目標(biāo)檢測/其他

【84】（anchor-free）Bridging the Gap Between Anchor-Based and Anchor-Free Detection via Adaptive Training Sample Selection

作者：Shifeng Zhang, Cheng Chi, Yongqiang Yao, Zhen Lei, Stan Z. Li

代碼：https://github.com/sfzhang15/ATSS

本文指出one-stage anchor-based和center-based anchor-free檢測算法間的差異主要來自于正負(fù)樣本的選擇，基于此提出ATSS(Adaptive Training Sample Selection)方法，該方法能夠自動根據(jù)GT的相關(guān)統(tǒng)計特征選擇合適的anchor box作為正樣本，在不帶來額外計算量和參數(shù)的情況下，能夠大幅提升模型的性能。

【85】（大規(guī)模/不均衡目標(biāo)檢測）Large-Scale Object Detection in the Wild From Imbalanced Multi-Labels

作者：Junran Peng, Xingyuan Bu, Ming Sun, Zhaoxiang Zhang, Tieniu Tan, Junjie Yan

【86】DLWL: Improving Detection for Lowshot Classes With Weakly Labelled Data

作者：Vignesh Ramanathan, Rui Wang, Dhruv Mahajan

【87】Correlation-Guided Attention for Corner Detection Based Visual Tracking

作者：Fei Du, Peng Liu, Wei Zhao, Xianglong Tang

【88】（特征檢測）Reinforced Feature Points: Optimizing Feature Detection and Description for a High-Level Task

作者：Aritra Bhowmik, Stefan Gumhold, Carsten Rother, Eric Brachmann

【89】Seeing Around Street Corners: Non-Line-of-Sight Detection and Tracking In-the-Wild Using Doppler Radar

作者：Nicolas Scheiner, Florian Kraus, Fangyin Wei, Buu Phan, Fahim Mannan, Nils Appenrodt, Werner Ritter, Jurgen Dickmann, Klaus Dietmayer, Bernhard Sick, Felix Heide

【90】Learning to Observe: Approximating Human Perceptual Thresholds for Detection of Suprathreshold Image Transformations

作者：Alan Dolhasz, Carlo Harvey, Ian Williams

【91】Siam R-CNN: Visual Tracking by Re-Detection

作者：Paul Voigtlaender, Jonathon Luiten, Philip H.S. Torr, Bastian Leibe

【92】Progressive Mirror Detection

作者：Jiaying Lin, Guodong Wang, Rynson W.H. Lau

【93】（陰影檢測）Instance Shadow Detection

作者：Tianyu Wang, Xiaowei Hu, Qiong Wang, Pheng-Ann Heng, Chi-Wing Fu

【94】（陰影檢測）A Multi-Task Mean Teacher for Semi-Supervised Shadow Detection

作者：Zhihao Chen, Lei Zhu, Liang Wan, Song Wang, Wei Feng, Pheng-Ann Heng

【95】（玻璃檢測）Don't Hit Me! Glass Detection in Real-World Scenes

作者：Haiyang Mei, Xin Yang, Yang Wang, Yuanyuan Liu, Shengfeng He, Qiang Zhang, Xiaopeng Wei, Rynson W.H. Lau

【96】Rethinking Classification and Localization for Object Detection

作者：Yue Wu, Yinpeng Chen, Lu Yuan, Zicheng Liu, Lijuan Wang, Hongzhi Li, Yun Fu

【97】（多anchor）Multiple Anchor Learning for Visual Object Detection

作者：Wei Ke, Tianliang Zhang, Zeyi Huang, Qixiang Ye, Jianzhuang Liu, Dong Huang

【98】Memory Enhanced Global-Local Aggregation for Video Object Detection

作者：Yihong Chen, Yue Cao, Han Hu, Liwei Wang 代碼：https://github.com/Scalsol/mega.pytorch

【99】CentripetalNet: Pursuing High-Quality Keypoint Pairs for Object Detection

作者：Zhiwei Dong, Guoxuan Li, Yue Liao, Fei Wang, Pengju Ren, Chen Qian

代碼：https://github.com/KiveeDong/CentripetalNet

本文提出一種使用向心偏移來對同一實例中的角點進行配對的CentripetalNet向心網(wǎng)絡(luò)。向心網(wǎng)絡(luò)可以預(yù)測角點的位置和向心偏移，并匹配移動結(jié)果對齊的角。結(jié)合位置信息，這種方法比傳統(tǒng)的嵌入方法更準(zhǔn)確地匹配角點。角池將邊界框內(nèi)的信息提取到邊界上。為了使這些信息在角落里更容易被察覺，作者又設(shè)計了一個交叉星可變形卷積網(wǎng)絡(luò)來適應(yīng)特征。除了檢測，通過為作者的CentripetalNet安置一個mask預(yù)測模塊來探索anchor-free檢測器上的實例分割。

【100】（one-stage）Learning From Noisy Anchors for One-Stage Object Detection作者：Hengduo Li, Zuxuan Wu, Chen Zhu, Caiming Xiong, Richard Socher, Larry S. Davis

【101】EfficientDet: Scalable and Efficient Object Detection

作者：Mingxing Tan, Ruoming Pang, Quoc V. Le

代碼：https://github.com/google/automl/tree/master/efficientdet

本文系統(tǒng)性地研究了多種檢測器架構(gòu)設(shè)計，試圖解決該問題?；趩坞A段檢測器范式，研究者查看了主干網(wǎng)絡(luò)、特征融合和邊界框/類別預(yù)測網(wǎng)絡(luò)的設(shè)計選擇，發(fā)現(xiàn)了兩大主要挑戰(zhàn)：高效的多尺度特征融合和模型縮放。針對這兩項挑戰(zhàn)，研究者提出了應(yīng)對方法：高效的多尺度特征融合和模型縮放。

【102】Overcoming Classifier Imbalance for Long-Tail Object Detection With Balanced Group Softmax

作者：Yu Li, Tao Wang, Bingyi Kang, Sheng Tang, Chunfeng Wang, Jintao Li, Jiashi Feng

【103】Dynamic Refinement Network for Oriented and Densely Packed Object Detection

作者：Xingjia Pan, Yuqiang Ren, Kekai Sheng, Weiming Dong, Haolei Yuan, Xiaowei Guo, Chongyang Ma, Changsheng Xu

代碼：https://github.com/Anymake/DRN_CVPR2020

【104】Noise-Aware Fully Webly Supervised Object Detection

作者：Yunhang Shen, Rongrong Ji, Zhiwei Chen, Xiaopeng Hong, Feng Zheng, Jianzhuang Liu, Mingliang Xu, Qi Tian

【105】Hit-Detector: Hierarchical Trinity Architecture Search for Object Detection

作者：Jianyuan Guo, Kai Han, Yunhe Wang, Chao Zhang, Zhaohui Yang, Han Wu, Xinghao Chen, Chang Xu

代碼：https://github.com/ggjy/HitDet.pytorch

【106】D2Det: Towards High Quality Object Detection and Instance Segmentation

作者：Jiale Cao, Hisham Cholakkal, Rao Muhammad Anwer, Fahad Shahbaz Khan, Yanwei Pang, Ling Shao

代碼：https://github.com/JialeCao001/D2Det

【107】Prime Sample Attention in Object Detection

作者：Yuhang Cao, Kai Chen, Chen Change Loy, Dahua Lin

【108】Exploring Categorical Regularization for Domain Adaptive Object Detection

作者：Chang-Dong Xu, Xing-Ran Zhao, Xin Jin, Xiu-Shen Wei

【109】SP-NAS: Serial-to-Parallel Backbone Search for Object Detection

作者：Chenhan Jiang, Hang Xu, Wei Zhang, Xiaodan Liang, Zhenguo Li

【110】NAS-FCOS: Fast Neural Architecture Search for Object Detection

作者：Ning Wang, Yang Gao, Hao Chen, Peng Wang, Zhi Tian, Chunhua Shen, Yanning Zhang

【111】Detection in Crowded Scenes: One Proposal, Multiple Predictions

作者：Xuangeng Chu, Anlin Zheng, Xiangyu Zhang, Jian Sun

代碼：https://github.com/megvii-model/CrowdDetection

【112】Cross-Domain Detection via Graph-Induced Prototype Alignment

作者：Minghao Xu, Hang Wang, Bingbing Ni, Qi Tian, Wenjun Zhang

【113】AugFPN: Improving Multi-Scale Feature Learning for Object Detection作者：Chaoxu Guo, Bin Fan, Qian Zhang, Shiming Xiang, Chunhong Pan

【114】Robust Object Detection Under Occlusion With Context-Aware CompositionalNets

作者：Angtian Wang, Yihong Sun, Adam Kortylewski, Alan L. Yuille

【115】（跨域目標(biāo)檢測）Cross-Domain Document Object Detection: Benchmark Suite and Method作者：Kai Li, Curtis Wigington, Chris Tensmeyer, Handong Zhao, Nikolaos Barmpalios, Vlad I. Morariu, Varun Manjunatha, Tong Sun, Yun Fu

【116】（跨域目標(biāo)檢測）Cross-domain Object Detection through Coarse-to-Fine Feature Adaptation

作者：Yangtao Zheng, Di Huang, Songtao Liu, Yunhong Wang

近年來，在基于深度學(xué)習(xí)的目標(biāo)檢測中見證了巨大的進步。但是，由于domain shift問題，將現(xiàn)成的檢測器應(yīng)用于未知的域會導(dǎo)致性能顯著下降。為了解決這個問題，本文提出了一種新穎的從粗到精的特征自適應(yīng)方法來進行跨域目標(biāo)檢測。由于這種從粗到細(xì)的特征自適應(yīng)，前景區(qū)域中的領(lǐng)域知識可以有效地傳遞。在各種跨域檢測方案中進行了廣泛的實驗，結(jié)果證明了所提出方法的廣泛適用性和有效性。

【117】Exploring Bottom-Up and Top-Down Cues With Attentive Learning for Webly Supervised Object Detection

作者：Zhonghua Wu, Qingyi Tao, Guosheng Lin, Jianfei Cai

【118】Context R-CNN: Long Term Temporal Context for Per-Camera Object Detection

作者：Sara Beery, Guanhang Wu, Vivek Rathod, Ronny Votel, Jonathan Huang

【119】Mixture Dense Regression for Object Detection and Human Pose Estimation

作者：Ali Varamesh, Tinne Tuytelaars

【120】Offset Bin Classification Network for Accurate Object Detection

作者：Heqian Qiu, Hongliang Li, Qingbo Wu, Hengcan Shi

【121】（Single Shot）NETNet: Neighbor Erasing and Transferring Network for Better Single Shot Object Detection

作者：Yazhao Li, Yanwei Pang, Jianbing Shen, Jiale Cao, Ling Shao

【122】Scale-Equalizing Pyramid Convolution for Object Detection

作者：Xinjiang Wang, Shilong Zhang, Zhuoran Yu, Litong Feng, Wayne Zhang

代碼：https://github.com/jshilong/SEPC

為了更好的解決物體檢測中的尺度問題，本文重新設(shè)計了經(jīng)典的單階段檢測器的FPN以及HEAD結(jié)構(gòu)，通過構(gòu)造更具等變性的特征金子塔，以提高檢測器應(yīng)對尺度變化的魯棒性，可以使單階段檢測器在coco上提升～4mAP。

【123】（邊界檢測）Joint Semantic Segmentation and Boundary Detection Using Iterative Pyramid Contexts

作者：Mingmin Zhen, Jinglu Wang, Lei Zhou, Shiwei Li, Tianwei Shen, Jiaxiang Shang, Tian Fang, Long Quan

【124】Physically Realizable Adversarial Examples for LiDAR Object Detection

作者：James Tu, Mengye Ren, Sivabalan Manivasagam, Ming Liang, Bin Yang, Richard Du, Frank Cheng, Raquel Urtasun

【125】Hierarchical Graph Attention Network for Visual Relationship Detection

作者：Li Mi, Zhenzhong Chen

【126】Training a Steerable CNN for Guidewire Detection

作者：Donghang Li, Adrian Barbu

【127】Deep Residual Flow for Out of Distribution Detection

作者：Ev Zisselman, Aviv tamar

【128】Cylindrical Convolutional Networks for Joint Object Detection and Viewpoint Estimation

作者：Sunghun Joung, Seungryong Kim, Hanjae Kim, Minsu Kim, Ig-Jae Kim, Junghyun Cho, Kwanghoon Sohn

【129】Learning a Unified Sample Weighting Network for Object Detection

作者：Qi Cai, Yingwei Pan, Yu Wang, Jingen Liu, Ting Yao, Tao Mei

【130】Seeing without Looking: Contextual Rescoring of Object Detections for AP Maximization

作者：Lourenco V. Pato, Renato Negrinho, Pedro M. Q. Aguiar

【131】（single stage）RetinaTrack: Online Single Stage Joint Detection and Tracking

作者：Zhichao Lu, Vivek Rathod, Ronny Votel, Jonathan Huang

【132】Universal Physical Camouflage Attacks on Object Detectors

作者：Lifeng Huang, Chengying Gao, Yuyin Zhou, Cihang Xie, Alan L. Yuille, Changqing Zou, Ning Liu

【133】SaccadeNet: A Fast and Accurate Object Detector

作者：Shiyi Lan, Zhou Ren, Yi Wu, Larry S. Davis, Gang Hua

【134】Generalized ODIN: Detecting Out-of-Distribution Image Without Learning From Out-of-Distribution Data

作者：Yen-Chang Hsu, Yilin Shen, Hongxia Jin, Zsolt Kira

【1357】A Programmatic and Semantic Approach to Explaining and Debugging Neural Network Based Object Detectors

作者：Edward Kim, Divya Gopinath, Corina Pasareanu, Sanjit A. Seshia

【136】Revisiting the Sibling Head in Object Detector

作者：Guanglu Song, Yu Liu, Xiaogang Wang

代碼：https://github.com/Sense-X/TSD

目前很多研究表明目標(biāo)檢測中的分類分支和定位分支存在較大的偏差，本文從sibling head改造入手，跳出常規(guī)的優(yōu)化方向，提出TSD方法解決混合任務(wù)帶來的內(nèi)在沖突，從主干的proposal中學(xué)習(xí)不同的task-aware proposal，同時結(jié)合PC來保證TSD的性能，在COCO上達到了51.2mAP。

獲取方式

在?極市平臺?后臺回復(fù)?CVPR檢測，即可獲取136篇論文打包下載鏈接。