刘文犀
 
基本信息
职称 副教授
职务 硕士生导师
主讲课程 ******
研究方向 计算机视觉
办公室 数计2号楼#404
电子邮件
联系电话
个人简介

刘文犀,副教授,硕士生导师,旗山学者计划。2015年加入计算机科学系认知系统与信息处理联合实验室。2014年获得香港城市大学计算机科学系博士学位。

研究方向:

  • 计算机视觉(人群分析、目标跟踪、图像分割、图像处理等);机器人(多机避障、导航等)等应用研究
  • 近期研究主要关注人群/交通/航拍场景下的图像视频分析(如检测、分割、跟踪等)

个人研究工作经历:

  • 2019.04 - 2020.07 香港大学计算机科学系博士后研究员(合作导师 Wenping Wang、Jia Pan)
  • 2014.10 - 2015.01 香港中文大学电子工程系研究助理(合作导师:Xiaogang Wang)
  • 2010.08 - 2014.10 香港城市大学计算机科学系博士(导师:Rynson W.H. Lau、Dinesh Manocha)
  • 2011.09 - 2012.02 北卡大学教堂山分校访问学者(合作导师:Dinesh Manocha)
  • 2010.04 - 2010.08 香港城市大学计算机科学系研究助理(合作导师:Rynson W.H. Lau)
  • 2009.09 - 2010.04 中国科学院深圳先进院客座学生(合作导师:Jianjun Ouyang和Lan Wang)
新闻
  • enlightened与华南理工大学合作论文《CrowdGAN: Identity-free Interactive Crowd Video Generation and Beyond》被人工智能顶级期刊Transactions on Pattern Analysis and Machine Intelligence(TPAMI)录用!
  • enlightened 与港大合作论文《Mapping in a Cycle: Sinkhorn Regularized Unsupervised Learning for Point Cloud Shapes》被计算机视觉顶级会议ECCV 2020录用!
  • enlightened 与港大合作论文《Distributed Multi-Robot Collision Avoidance via Deep Reinforcement Learning for Navigation in Complex Scenarios》被机器人学顶级期刊IJRR录用!
  • enlightened 本组论文论文《Visualizing the Invisible: Occluded Vehicle Segmentation and Recovery》被计算机视觉顶级会议ICCV 2019录用! [福大要闻]
  • enlightened 本组论文论文《Context-aware Spatio-recurrent Curvilinear Structure Segmentation》被计算机视觉顶级会议CVPR 2019录用![福大要闻] 
部分发表论文
  • Liangyu Chai, Yongtuo Liu, Wenxi Liu, Guoqiang Han, Shengfeng He: CrowdGAN: Identity-free Interactive Crowd Video Generation and Beyond. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2020
  • Yongtuo Liu, Qiang Wen, Haoxin Chen, Wenxi Liu, Jing Qin, Guoqiang Han, Shengfeng He: Crowd Counting via Cross-stage Refinement Networks. IEEE Transactions on Image Processing (TIP), 2020
  • Yuzhen Niu, Guanchao Long, Wenxi Liu, Wenzhong Guo, Shengfeng He: Boundary-aware RGBD Salient Object Detection with Cross-modal Feature Sampling, IEEE Transactions on Image Processing (TIP), 2020
  • Bo Wang, Yiliang Chen, Wenxi Liu, Jing Qin, Yong Du, Guoqiang Han, Shengfeng He: Real-Time Hierarchical Supervoxel Segmentation via a Minimum Spanning Tree, IEEE Transactions on Image Processing (TIP), 2020
  • Bo Wang, Wenxi Liu, Guoqiang Han, Shengfeng He: Learning Long-term Structural Dependencies for Video Salient Object Detection, IEEE Transactions on Image Processing (TIP), 2020
  • Yuzhen Niu, Qingyang Zheng, Wenxi Liu, Wenzhong Guo: Recurrent Enhancement of Visual Comfort for Casual Stereoscopic Photography, IEEE Virtual Reality (VR) 2020 (Oral Presentation)
  • Tingxiang Fan, Pinxin Long, Wenxi Liu, Jia Pan: Distributed Multi-Robot Collision Avoidance via Deep Reinforcement Learning for Navigation in Complex Scenarios. I. J. Robotics Res. (IJRR) 2020
  • Lei Yang, Wenxi Liu, Zhiming Cui, Nenglun Chen, Wenping Wang: Mapping in a Cycle: Sinkhorn Regularized Unsupervised Learning for Point Cloud Shapes. ECCV 2020
  • Chun-Yang Zhang, Yongyi Xiao, Jin-Cheng Lin, C. L. Philip Chen, Wenxi Liu, Yuhong Tong: 3D Deconvolutional Networks for the Unsupervised Representation Learning of Human Motions, IEEE Transactions on Cybernetics 2020.
  • Tao Yan, Jianbo Jiao, Wenxi Liu, Rynson WH Lau: Stereoscopic Image Generation from Light Field with Disparity Scaling and Super-Resolution IEEE Trans. Image Processing (TIP), 2020
  • Tao Yan, Yiming Mao, Jianming Wang, Wenxi Liu, Xiaohua Qian, and Rynson Lau: Generating Stereoscopic Images with Convergence Control Ability from a Light Field Image Pair. IEEE Trans. Circuits Syst. Video Techn. (TCSVT) 2020
  • Wenxi Liu, Chun-Yang Zhang, Genggeng Liu, Yaru Su, Naixue Xiong: Extraversion Measure for Crowd Trajectories. IEEE Trans. on Industrial Informatics (TII), Special Section on Emerging Trends Issues and Challenges in Edge Artificial Intelligence, 2019
  • Xiaosheng Yan, Feigege Wang, Wenxi Liu*, Yuanlong Yu*, Shengfeng He, Jia Pan: Visualizing the Invisible: Occluded Vehicle Segmentation and Recovery. ICCV 2019
  • Feigege Wang, Yue Gu, Wenxi Liu*, Yuanlong Yu, Shengfeng He, Jia Pan: Context-aware Spatio-recurrent Curvilinear Structure Segmentation. CVPR 2019
  • Wenxi Liu, Yibing Song, Dengsheng Chen, Shengfeng He, Yuanlong Yu, Tao Yan, Rynson W. H. Lau: Deformable Object Tracking with Gated Fusion. IEEE Transactions on Image Processing (TIP), 2019
  • Pinxin Long, Tingxiang Fan, Xinyi Liao, Wenxi Liu, Hao Zhang, Jia Pan: Towards Optimally Decentralized Multi-Robot Collision Avoidance via Deep Reinforcement Learning. ICRA 2018
  • Pinxin Long, Wenxi Liu, Jia Pan: Deep-Learned Collision Avoidance Policy for Distributed Multiagent Navigation. IEEE Robotics and Automation Letters 2(2): 656-663, 2017 [Presented at ICRA 2017]
  • Wenxi Liu, Rynson W. H. Lau, Xiaogang Wang, Dinesh Manocha: Exemplar-AMMs: Recognizing Crowd Movements From Pedestrian Trajectories. IEEE Trans. Multimedia (TMM) 18(12): 2398-240, 2016 [ PDF ]
  • Wenxi Liu, Rynson W. H. Lau, Dinesh Manocha: Robust individual and holistic features for crowd scene classification. Pattern Recognition (PR) 58: 110-120, 2016 [ PDF ]
  • Shengfeng He, Rynson W. H. Lau, Wenxi Liu, Zhe Huang, Qingxiong Yang: SuperCNN: A Superpixelwise Convolutional Neural Network for Salient Object Detection. International Journal of Computer Vision (IJCV) 115(3): 330-344, 2015
  • Sujeong Kim, Stephen J. Guy, Wenxi Liu, David Wilkie, Rynson W. H. Lau, Ming C. Lin, Dinesh Manocha: BRVO: Predicting pedestrian trajectories using velocity-space reasoning. I. J. Robotics Res. (IJRR) 34(2): 201-217, 2015
  • Wenxi Liu, Antoni B. Chan, Rynson W. H. Lau, Dinesh Manocha: Leveraging Long-Term Predictions and Online Learning in Agent-Based Multiple Person Tracking. IEEE Trans. Circuits Syst. Video Techn. (TCSVT) 25(3): 399-410, 2015 [ PDF ]
  • Stephen J. Guy, Jur van den Berg, Wenxi Liu, Rynson W. H. Lau, Ming C. Lin, Dinesh Manocha: A statistical similarity measure for aggregate crowd dynamics. ACM Trans. Graph. (TOG) 31(6): 190:1-190:11, 2012 [ Presented at SIGGRAPH Asia 2012 | Project Page ]
  • Wenxi Liu, Rynson W. H. Lau, Dinesh Manocha: Crowd simulation using Discrete Choice Model. IEEE Virtual Reality (VR) 2012: 3-6
招生要求
  • 我们组做什么?有趣的、有创意的、有质量的科研探索
  • 目标是什么?一般来说目标为国际顶级会议或期刊论文,但非必要条件
  • 我们组能提供什么?从入门到完成科研项目/论文的一对一指导;本组有在AI顶级会议期刊上发表论文的丰富经验,并与国内外优秀科研团队有密切合作
  • 对学生有什么要求?Self-motivated and Persistent
    • enlightened在此基础上,获得保研资格、有程序设计竞赛经验或有深度学习项目经验的同学优先考虑
    • enlightened编程基本功扎实、具有一定机器学习基础的本科同学们(任意年级任意时间)可联系我尝试【入门项目】,以决定未来方向
    • 关于更详细的信息,欢迎同学们发送Email(wenxi.liu@hotmail.com)咨询