(根據研究方向簡單分類,供參考,後續會根據大家的意見改進分類,看完後有點幫助的話,還煩請掃描文末二維碼,關注同步的公眾號( ̄▽ ̄)")

cs.CV 方向,今日共計55篇[檢測分類相關]: 【1】 FireNet: A Specialized Lightweight Fire & Smoke Detection Model for Real-Time IoT ApplicationsFireNet:用於實時物聯網應用的專業輕量級火災和煙霧探測模型

作者: Arpit Jadon, Rishabh Sharma

備註:Under review in IEEE Region 10 Conference (TENCON) 2019, Paper ID 1307, Track T03.2: Neural Networks and Deep Learning鏈接:arxiv.org/abs/1905.1192【2】 Compositional Convolutional Networks For Robust Object Classification under Occlusion遮擋下魯棒目標分類的組合卷積網路作者: Adam Kortylewski, Alan Yuille 鏈接:arxiv.org/abs/1905.1182【3】 Hallucinating Optical Flow Features for Video Classification用於視頻分類的幻覺光流特徵作者: Yongyi Tang, Lianqiang Zhou

備註:Accepted by IJCAI 2019

鏈接:arxiv.org/abs/1905.1179【4】 Integrated Neural Network and Machine Vision Approach For Leather Defect Classification皮革缺陷分類的集成神經網路與機器視覺方法作者: Sze-Teng Liong, Wei-Chuen Yau 鏈接:arxiv.org/abs/1905.1173【5】 Union Visual Translation Embedding for Visual Relationship Detection and Scene Graph Generation用於視覺關係檢測和場景圖生成的聯合視覺翻譯嵌入作者: Zih-Siou Hung, Svetlana Lazebnik 鏈接:arxiv.org/abs/1905.1162

【6】 Jointly Learning Structured Analysis Discriminative Dictionary and Analysis Multiclass Classifier

共同學習結構化分析判別詞典和分析多類分類器作者: Zhao Zhang, Shuicheng Yan 鏈接:arxiv.org/abs/1905.1154[行為/時空/光流/姿態/運動]: 【1】 Importance of user inputs while using incremental learning to personalize human activity recognition models使用增量學習來個性化人類活動識別模型時用戶輸入的重要性作者: Pekka Siirtola, Juha R?ning 備註:European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN) 2019, pages 449-454鏈接:arxiv.org/abs/1905.1177

[分割相關]:

【1】 Road Segmentation with Image-LiDAR Data Fusion利用Image-LiDAR數據融合進行道路分割作者: Huafeng Liu, Zhenmin Tang 鏈接:arxiv.org/abs/1905.1155【2】 Enhancing Salient Object Segmentation Through Attention通過注意增強顯著對象分割作者: Anuj Pahuja, R. Venkatesh Babu 備註:CVPRW - Deep Vision 2019鏈接:arxiv.org/abs/1905.1152

[GAN/對抗式學習相關]:

【1】 Cross-Domain Transferability of Adversarial Perturbations對抗擾動的跨域可轉移性作者: Muzammal Naseer, Fatih Porikli 鏈接:arxiv.org/abs/1905.1173【2】 JGAN: A Joint Formulation of GAN for Synthesizing Images and LabelsJGAN:GAN用於合成圖像和標籤的聯合配方作者: Minje Park 鏈接:arxiv.org/abs/1905.1157【3】 ME-Net: Towards Effective Adversarial Robustness with Matrix Estimation

ME-Net:利用矩陣估計實現有效的對抗魯棒性

作者: Yuzhe Yang, Zhi Xu 備註:ICML 2019鏈接:arxiv.org/abs/1905.1197【4】 Adversarial Domain Adaptation Being Aware of Class Relationships對抗性領域適應意識到階級關係作者: Zeya Wang, Eric P. Xing 鏈接:arxiv.org/abs/1905.1193【5】 Scaleable input gradient regularization for adversarial robustness可擴展的輸入梯度正則化,用於對抗魯棒性

作者: Chris Finlay, Adam M Oberman

鏈接:arxiv.org/abs/1905.1146【6】 Trust but Verify: An Information-Theoretic Explanation for the Adversarial Fragility of Machine Learning Systems, and a General Defense against Adversarial Attacks信任但驗證:機器學習系統的對抗性脆弱性的信息理論解釋,以及針對對抗性攻擊的一般防禦作者: Jirong Yi, Raghuraman Mudumbai 備註:44 Pages, 2 Theorems, 35 Figures, 29 Tables. arXiv admin note: substantial text overlap with arXiv:1901.09413鏈接:arxiv.org/abs/1905.1138[半/弱/無監督相關]: 【1】 Unsupervised Learning from Video with Deep Neural Embeddings基於深度神經嵌入的視頻無監督學習

作者: Chengxu Zhuang, Daniel Yamins

鏈接:arxiv.org/abs/1905.1195【2】 SizeNet: Weakly Supervised Learning of Visual Size and Fit in Fashion ImagesSizeNet:視覺大小的弱監督學習和適合時尚形象作者: Nour Karessli, Reza Shirvany 備註:IEEE Conference on Computer Vision and Pattern Recognition Workshop (CVPRW) 2019 Focus on Fashion and Subjective Search - Understanding Subjective Attributes of Data (FFSS-USAD)鏈接:arxiv.org/abs/1905.1178【3】 Local Label Propagation for Large-Scale Semi-Supervised Learning大規模半監督學習的局部標籤傳播作者: Chengxu Zhuang, Daniel Yamins

鏈接:arxiv.org/abs/1905.1158

[裁剪/量化/加速相關]: 【1】 Online Filter Clustering and Pruning for Efficient Convnets在線過濾器聚類和修剪有效的Convnet作者: Zhengguang Zhou, Houqiang Li 鏈接:arxiv.org/abs/1905.1178【2】 CGaP: Continuous Growth and Pruning for Efficient Deep LearningCGaP:持續增長和修剪有效的深度學習作者: Xiaocong Du, Yu Cao 鏈接:arxiv.org/abs/1905.1153【3】 Differentiable Quantization of Deep Neural Networks深度神經網路的可微量化作者: Stefan Uhlich, Akira Nakamura 鏈接:arxiv.org/abs/1905.1145【4】 Additive Noise Annealing and Approximation Properties of Quantized Neural Networks量化神經網路的加性雜訊??退火和逼近性質作者: Matteo Spallanzani, Luca Benini 鏈接:arxiv.org/abs/1905.1045[Re-id相關]: 【1】 Video-based Person Re-identification with Two-stream Convolutional Network and Co-attentive Snippet Embedding基於視頻的人物重新識別與雙流卷積網路和共同注意的片段嵌入作者: Peixian Chen, Yuyu Huang 鏈接:arxiv.org/abs/1905.1186[其他]: 【1】 Cerberus: A Multi-headed DerendererCerberus:一個多頭的Derenderer作者: Boyang Deng, Geoffrey Hinton 鏈接:arxiv.org/abs/1905.1194【2】 An Analysis of Object Embeddings for Image Retrieval圖像檢索的對象嵌入分析作者: Bor-Chun Chen, Ser-Nam Lim 鏈接:arxiv.org/abs/1905.1190【3】 FaceSwapNet: Landmark Guided Many-to-Many Face ReenactmentFaceSwapNet:具有里程碑意義的引導多對多面部再現作者: Jiangning Zhang, Changjie Fan 鏈接:arxiv.org/abs/1905.1180【4】 Progressive Learning of Low-Precision Networks低精度網路的漸進式學習作者: Zhengguang Zhou, Houqiang Li 鏈接:arxiv.org/abs/1905.1178【5】 PHT-bot: Deep-Learning based system for automatic risk stratification of COPD patients based upon signs of Pulmonary HypertensionPHT-bot:基於深度學習的系統,用於根據肺動脈高壓的體征對COPD患者進行自動風險分層作者: David Chettrit, Eldad Elnekave 鏈接:arxiv.org/abs/1905.1177【6】 A Cost Efficient Approach to Correct OCR Errors in Large Document Collections一種經濟有效的方法來糾正大型文檔集合中的OCR錯誤作者: Deepayan Das, C. V. Jawahar 鏈接:arxiv.org/abs/1905.1173【7】 Invertible generative models for inverse problems: mitigating representation error and dataset bias逆問題的可逆生成模型:減輕表示誤差和數據集偏差作者: Muhammad Asim, Paul Hand 鏈接:arxiv.org/abs/1905.1167【8】 OICSR: Out-In-Channel Sparsity Regularization for Compact Deep Neural NetworksOICSR:緊湊型深度神經網路的信道外稀疏正則化作者: Jiashi Li, Haifeng Sun 備註:Accepted to CVPR 2019鏈接:arxiv.org/abs/1905.1166【9】 The Nipple-Areola Complex for Criminal Identification用於刑事鑒定的乳頭 - 乳暈複合體作者: Wojciech Michal Matkowski, Cory Lloyd Hall 備註:Accepted in the International Conference on Biometrics (ICB 2019), scheduled for 4-7 June 2019 in Crete, Greece鏈接:arxiv.org/abs/1905.1165【10】 Image Deformation Meta-Networks for One-Shot Learning用於一次性學習的圖像變形元網路作者: Zitian Chen, Martial Hebert 備註:Oral at CVPR2019鏈接:arxiv.org/abs/1905.1164【11】 LatentGNN: Learning Efficient Non-local Relations for Visual RecognitionLatentGNN:學習有效的非局部關係進行視覺識別作者: Songyang Zhang, Xuming He 備註:ICML 2019鏈接:arxiv.org/abs/1905.1163【12】 Improving Action Localization by Progressive Cross-stream Cooperation通過漸進式跨流合作改進行動本地化作者: Rui Su, Dong Xu 備註:CVPR2019鏈接:arxiv.org/abs/1905.1157【13】 Case-Based Histopathological Malignancy Diagnosis using Convolutional Neural Networks基於病例的組織病理學惡性腫瘤診斷使用卷積神經網路作者: Qicheng Lao, Thomas Fevens 鏈接:arxiv.org/abs/1905.1156【14】 Semantic Fisher Scores for Task Transfer: Using Objects to Classify Scenes任務轉移的語義Fisher分數:使用對象對場景進行分類作者: Mandar Dixit, Nuno Vasconcelos 鏈接:arxiv.org/abs/1905.1153【15】 Shape Evasion: Preventing Body Shape Inference of Multi-Stage Approaches形狀規避:防止多階段方法的體形推斷作者: Hosnieh Sattar, Mario Fritz 鏈接:arxiv.org/abs/1905.1150【16】 End-to-End Pore Extraction and Matching in Latent Fingerprints: Going Beyond Minutiae潛在指紋中的端到端孔隙提取和匹配:超越細節作者: Dinh-Luan Nguyen, Anil K. Jain 鏈接:arxiv.org/abs/1905.1147【17】 A Symmetric Encoder-Decoder with Residual Block for Infrared and Visible Image Fusion一種具有殘差塊的對稱編碼器 - 解碼器,用於紅外和可見光圖像融合作者: Lihua Jian, David Chisholm 鏈接:arxiv.org/abs/1905.1144【18】 EfficientNet: Rethinking Model Scaling for Convolutional Neural NetworksEfficientNet:重新思考卷積神經網路的模型尺度作者: Mingxing Tan, Quoc V. Le 備註:Published in ICML 2019鏈接:arxiv.org/abs/1905.1194【19】 A Compact Representation of Histopathology Images using Digital Stain Separation & Frequency-Based Encoded Local Projections使用數字染色分離和基於頻率的編碼局部投影的緊湊的組織病理學圖像表示作者: Alison K. Cheeseman, Edward R. Vrscay 備註:Accepted for publication in the International Conference on Image Analysis and Recognition (ICIAR 2019)鏈接:arxiv.org/abs/1905.1194【20】 Network Deconvolution網路解卷積作者: Chengxi Ye, Yiannis Aloimonos 鏈接:arxiv.org/abs/1905.1192【21】 BreizhCrops: A Satellite Time Series Dataset for Crop Type IdentificationBreizhCrops:用於作物類型識別的衛星時間序列數據集作者: Marc Ru?wurm, Marco K?rner 鏈接:arxiv.org/abs/1905.1189【22】 Snooping Attacks on Deep Reinforcement Learning窺探深層強化學習的攻擊作者: Matthew Inkawhich, Hai Li 鏈接:arxiv.org/abs/1905.1183【23】 Deep Scale-spaces: Equivariance Over Scale深度尺度空間:超越尺度的等效性作者: Daniel E. Worrall, Max Welling 鏈接:arxiv.org/abs/1905.1169【24】 Learning Dynamics of Attention: Human Prior for Interpretable Machine Reasoning學習注意力動機:人類在可解釋的機器推理之前作者: Wonjae Kim, Yoonho Lee 鏈接:arxiv.org/abs/1905.1166【25】 Discrete Infomax Codes for Meta-Learning用於元學習的離散Infomax代碼作者: Yoonho Lee, Seungjin Choi 鏈接:arxiv.org/abs/1905.1165【26】 Adaptive Lighting for Data-Driven Non-Line-of-Sight 3D Localization and Object Identification用於數據驅動的非視距3D定位和對象識別的自適應照明作者: Sreenithy Chandran, Suren Jayasuriya 鏈接:arxiv.org/abs/1905.1159【27】 Label Universal Targeted Attack標籤通用目標攻擊作者: Naveed Akhtar, Ajmal Mian 鏈接:arxiv.org/abs/1905.1154【28】 Improved Training Speed, Accuracy, and Data Utilization Through Loss Function Optimization通過損失函數優化提高訓練速度,準確性和數據利用率作者: Santiago Gonzalez, Risto Miikkulainen 鏈接:arxiv.org/abs/1905.1152【29】 FAN: Focused Attention Networks范:聚焦注意力網路作者: Chu Wang, Kaleem Siddiqi 鏈接:arxiv.org/abs/1905.1149【30】 Capsule Routing via Variational Bayes通過變分貝葉斯的膠囊路由作者: Fabio De Sousa Ribeiro, Stefanos Kollias 鏈接:arxiv.org/abs/1905.1145【31】 Equivalent and Approximate Transformations of Deep Neural Networks深度神經網路的等價和近似變換作者: Abhinav Kumar, Srikumar Ramalingam 鏈接:arxiv.org/abs/1905.1142【32】 Automatic Delineation of Kidney Region in DCE-MRIDCE-MRI中腎臟區域的自動劃分作者: Santosh Tirunagari, David Windridge 備註:arXiv admin note: text overlap with arXiv:1905.10218鏈接:arxiv.org/abs/1905.1138翻譯:谷歌翻譯

推薦閱讀:
相关文章