谷歌CVPR最全总结：45篇论文，Ian Goodfellow GAN演讲PPT下载

原标题：谷歌CVPR最全总结：45篇论文，Ian Goodfellow GAN演讲PPT下载新智元 - CVPR 2018专题来源：Google、iangoodfellow.com整理：肖琴【新智元导读】谷歌在今年的CVPR上表现强势，有超过200名谷歌员工将在大会上展示论文或被邀请演讲，45篇论文被接收。在计算机视觉领域，生成对抗网络GAN无疑是最受关注的主题之一，本文一并带来谷歌 Staff Research Scientist、GAN的提出人Ian Goodfellow在CVPR 2018上作关于GAN的演讲的PPT。地址：https://ai.googleblog.comhttp://www.iangoodfellow.com/slides/2018-06-18.pdf今天，2018年计算机视觉和模式识别会议（CVPR 2018）正在盐湖城举办，这是计算机视觉领域最重要的年度学术会议，包括主大会和若干workshop和tutorial。作为会议的钻石赞助商，谷歌在今年的CVPR上同样表现强势，有超过200名谷歌员工将在大会上展示论文或被邀请演讲，谷歌也组织和参与了多个研讨会。根据谷歌官方博客，CVPR 2018谷歌共有45篇论文被接收。这些论文关注下一代智能系统和机器感知领域的最新机器学习技术，包括Pixel 2和Pixel 2 XL智能手机的人像模式背后的技术，V4版本的Open Images数据集等等。Google at CVPR 2018组织者论文列表作为结构表示的对象标志的无监督发现Unsupervised Discovery of Object Landmarks as Structural RepresentationsYuting Zhang, Yijie Guo, Yixin Jin, Yijun Luo, Zhiyuan He, Honglak LeeDoubleFusion：利用单个深度传感器实时捕捉人体的内体形状DoubleFusion: Real-time Capture of Human Performances with Inner Body Shapes from a Single Depth SensorTao Yu, Zerong Zheng, Kaiwen Guo, Jianhui Zhao, Qionghai Dai, Hao Li, Gerard Pons-Moll, Yebin Liu用于无监督运动重定向的神经运动网络Neural Kinematic Networks for Unsupervised Motion RetargettingRuben Villegas, Jimei Yang, Duygu Ceylan, Honglak Lee用核预测网络去噪Burst Denoising with Kernel Prediction NetworksBen Mildenhall, Jiawen Chen, Jonathan Barron, Robert Carroll, Dillon Sharlet, Ren Ng神经网络的量化和训练，以实现高效的整数运算推理Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only InferenceBenoit Jacob, Skirmantas Kligys, Bo Chen, Matthew Tang, Menglong Zhu, Andrew Howard, Dmitry Kalenichenko, Hartwig AdamAVA：一个时空本地化原子视觉动作视频数据集AVA: A Video Dataset of Spatio-temporally Localized Atomic Visual ActionsChunhui Gu, Chen Sun, David Ross, Carl Vondrick, Caroline Pantofaru, Yeqing Li, Sudheendra Vijayanarasimhan, George Toderici, Susanna Ricco, Rahul Sukthankar, Cordelia Schmid, Jitendra Malik视觉问答的视觉-文本注意力焦点Focal Visual-Text Attention for Visual Question AnsweringJunwei Liang, Lu Jiang, Liangliang Cao, Li-Jia Li, Alexander G. Hauptmann推断来自阴影中的光场Inferring Light Fields from ShadowsManel Baradad, Vickie Ye, Adam Yedida, Fredo Durand, William Freeman, Gregory Wornell, Antonio Torralba修改多个视图中的非本地变量Modifying Non-Local Variations Across Multiple ViewsTal Tlusty, Tomer Michaeli, Tali Dekel, Lihi Zelnik-Manor超越卷积的迭代视觉推理Iterative Visual Reasoning Beyond ConvolutionsXinlei Chen, Li-jia Li, Fei-Fei Li, Abhinav Gupta3D形变模型回归的无监督训练Unsupervised Training for 3D Morphable Model RegressionKyle Genova, Forrester Cole, Aaron Maschinot, Daniel Vlasic, Aaron Sarna, William Freeman学习可扩展图像识别的可转换架构Learning Transferable Architectures for Scalable Image RecognitionBarret Zoph, Vijay Vasudevan, Jonathon Shlens, Quoc Le生物物种分类和检测数据集The iNaturalist Species Classification and Detection DatasetGrant van Horn, Oisin Mac Aodha, Yang Song, Yin Cui, Chen Sun, Alex Shepard, Hartwig Adam, Pietro Perona, Serge Belongie利用观察世界来学习内在的图像分解Learning Intrinsic Image Decomposition from Watching the WorldZhengqi Li, Noah Snavely学习智能对话框用于边界框注释Learning Intelligent Dialogs for Bounding Box AnnotationKsenia Konyushkova, Jasper Uijlings, Christoph Lampert, Vittorio Ferrari重新审视训练对象类别检测器的知识迁移Revisiting Knowledge Transfer for Training Object Class DetectorsJasper Uijlings, Stefan Popov, Vittorio Ferrari重新思考用Faster R-CNN架构进行时间动作定位Rethinking the Faster R-CNN Architecture for Temporal Action LocalizationYu-Wei Chao, Sudheendra Vijayanarasimhan, Bryan Seybold, David Ross, Jia Deng, Rahul Sukthankar视觉对象识别的层次式新颖性检测Hierarchical Novelty Detection for Visual Object RecognitionKibok Lee, Kimin Lee, Kyle Min, Yuting Zhang, Jinwoo Shin, Honglak LeeCOCO-Stuff：语境中的事物和材料类别COCO-Stuff: Thing and Stuff Classes in ContextHolger Caesar, Jasper Uijlings, Vittorio Ferrari用于视频分类的外观关系网络Appearance-and-Relation Networks for Video ClassificationLimin Wang, Wei Li, Wen Li, Luc Van GoolMorphNet：深度网络的快速简单资源约束结构学习MorphNet: Fast & Simple Resource-Constrained Structure Learning of Deep NetworksAriel Gordon, Elad Eban, Bo Chen, Ofir Nachum, Tien-Ju Yang, Edward Choi图形卷积自动编码器的可变形形状补完Deformable Shape Completion with Graph Convolutional AutoencodersOr Litany, Alex Bronstein, Michael Bronstein, Ameesh MakadiaMegaDepth：从互联网照片学习单视图深度预测MegaDepth: Learning Single-View Depth Prediction from Internet PhotosZhengqi Li, Noah Snavely作为结构表示的对象标志的无监督发现Unsupervised Discovery of Object Landmarks as S