| 为促进国内计算机视觉研究的发展，加强工业界、学术界的交流，举办CVPR 2018中国论文宣讲研讨会。本次活动由微软亚洲研究院主办，清华大学媒体与网络技术教育部-微软重点实验室、商汤科技承办，中国计算机学会计算机视觉专委会、中国图象图形学会视觉大数据专委会协办。 来自国内外计算机视觉领域学术界、工业界的优秀代表们将携各自在CVPR 2018发表的最新研究和技术观点，在此次论坛上与大家分享、交流、讨论！ |
| 8:40am-9:00am | 签到入场 | |
| 开幕 | 王井东- 微软亚洲研究院资深研究员，清华-微软联合实验室微软方副主任 | |
| 9:00am-9:10am | 马歆- 微软亚洲研究院学术合作部总监 | |
Session 1: GAN and Synthesis. Chair: 张婷 | ||
| 9:10am-9:25am | 刘家瑛 - 北京大学 | Attentive Generative Adversarial Network for Raindrop Removal from A Single Image |
| 9:25am-9:40am | 胡一博 - 中科院自动化所 | Pose-Guided Photorealistic Face Rotation |
| 9:40am-9:55am | 连宙辉 - 北京大学 | A Common Framework for Interactive Texture Transfer |
| 9:55am-10:10am | 傅建龙 - 微软亚洲研究院 | DA-GAN: Instance-level Image Translation by Deep Attention Generative Adversarial Network |
Session 2: Deep Learning. Chair: 刘家瑛 | ||
| 10:10am-10:25am | 张婷 - 微软亚洲研究院 | Interleaved Group Convolutions V2 |
| 10:25am-10:40am | 苏航 - 清华大学 | Interpret Neural Networks by Identifying Critical Data Routing Paths | 10:40am-10:55am | 龙明盛 - 清华大学 | Partial Transfer Learning with Selective Adversarial Networks |
| 10:55am-11:10am | 李培华 - 大连理工大学 | Towards Faster Training of Global Covariance Pooling Networks by Iterative Matrix Square Root Normalization |
| 11:10am-12:00pm | 凌海滨 - Temple University 王晓刚 - 香港中文大学 何旭明 - 上海科技大学 朱军 - 清华大学 王井东 - 微软亚洲研究院 陈光 - 爱可可爱生活 - 北京邮电大学 | 圆桌讨论 主持人：刘偲 - 中国科学院信息工程研究所 |
| 12:00pm-1:30pm | 午餐 + 海报展示 | |
Session 3: Person Re-Identification and Tracking. Chair: 傅建龙 | ||
| 1:30pm-1:45pm | 王乃岩 - 图森未来 | Multi-shot pedestrian re-identification via sequential decision making |
| 1:45pm-2:00pm | 张史梁 - 北京大学 | Person Transfer GAN to Bridge Domain Gap for Person Re-Identification |
| 2:00pm-2:15pm | 李鸿升 - 港中文-商汤联合实验室 | Towards More Robust Person Re-identification with Group Consistency and Background-bias Elimination |
| 2:15pm-2:30pm | 武伟 - 商汤科技 | High Performance Object Tracking with Siamese-network |
Session 4: Vision and Language. Chair: 石建萍 | ||
| 2:30pm-2:45pm | 段楠 - 微软亚洲研究院 | Visual Question Generation as Dual Task of Visual Question Answering |
| 2:45pm-3:00pm | 黄岩 - 中科院自动化所 | Learning Semantic Concepts and Order for Image and Sentence Matching |
| 3:00pm-3:15pm | 王鹏 - 西北工业大学 | Are You Talking to Me? Reasoned Visual Dialog Generation through Adversarial Learning |
| 3:15pm-3:30pm | 谭明奎 - 华南理工大学 | Visual grounding via accumulated attention |
| 3:30pm-4:30pm | 茶歇 + 海报展示 | |
Session 5: Segmentation, Detection. Chair: 谭明奎 | 4:30pm-4:45pm | 戴玉超 - 西北工业大学 | Deep Unsupervised Saliency Detection: A Multiple Noisy Labeling Perspective |
| 4:45pm-5:00pm | 石建萍 - 商汤科技 | Effective Scene Understanding for Autonomous Driving |
| 5:00pm-5:15pm | 廖菁 - 微软亚洲研究院 | Stereoscopic Neural Style Transfer |
Session 6: Human, Face and 3D Shape. Chair: 廖菁 | ||
| 5:15pm-5:30pm | 卢策吾 - 上海交通大学 | Weakly-supervised Human body Part parsing via pose guided knowledge transfer |
| 5:30pm-5:45pm | 赵凯莉 - 北京邮电大学 | Learning Facial Action Units from Web Images with Scalable Weakly Supervised Clustering |
| 5:45pm-6:00pm | 高跃 - 清华大学 | GVCNN: Group-View Convolutional Neural Networks for 3D Shape Recognition |
海报展示 | |
| 折栋宇 - 南开大学 | Weakly Supervised Coupled Networks for Visual Sentiment Analysis |
| 陈章 - 上海科技大学 | Sparse Photometric 3D Face Reconstruction Guided by Morphable Models |
| 杨鸿宇 - 北京航空航天大学 | Learning Face Age Progression: A Pyramid Architecture of GANs |
| 王晓迪 - 北京航空航天大学 | Modulated Convolutional Network |
| 杨一博 - 北京大学 | Convolutional Neural Networks with Alternately Updated Clique |
| 黄雷 - 北京航空航天大学 | Decorrelated Batch Normalization |
| 万纬韬 - 清华大学 | Rethinking Feature Distribution for Loss Functions in Image Classification |
| 黄鑫 - 北京大学 | Deep Cross-media Knowledge Transfer |
| 曹越 - 清华大学 | Deep Cauchy Hashing for Hamming Space Retrieval |
| 黄永业 - 北京邮电大学 | SketchMate: Deep Hashing for Million-Scale Human Sketch Retrieval |
| 杨韧 - 北京航空航天大学 | Multi-Frame Quality Enhancement for Compressed Video |
| 李明晗 - 西安交通大学 | Video Rain Streak Removal By Multiscale Sparse CODing |
| 郁环宇 - 上海交通大学 | Fine-grained video captioning for sports narrative |
| 陈恺 - 港中文-商汤联合实验室 | Optimizing Video Object Detection via a Scale-Time Lattice |
| 张士峰 - 中国科学院自动化研究所 | Single-Shot Refinement Neural Network for Object Detection |
| 四建楼 - 北京邮电大学 | Dual Attention Matching Network for Context-Aware Feature Sequence based Person Re-Identification |
| 黄青虬 - 港中文-商汤联合实验室 | Unifying Identification and Context Learning for Person Recognition |
| 孙晓晓 - 南开大学 | Clinical Skin Lesion Diagnosis using Representations Inspired by Dermatologist Criteria |
| 姚太平 - 上海交通大学 | Multiple Granularity Group Interaction Prediction |
| 庞家昊 - 商汤科技 | Zoom and Learn: Generalizing Deep Stereo Matching to Novel Domains |
| 殷志超 - 商汤科技 | GeoNet: Unsupervised Learning of Dense Depth, Optical Flow and Camera Pose |
| 刘永 - 中科院计算所 | Structure Inference Net: Object Detection Using Scene-Level Context and Instance-Level Relationships |
| 黄子龙 - 华中科技大学 | Weakly-Supervised Semantic Segmentation Network with Deep Seeded Region Growing |