Yuanlu Xu's Homepage

Marching with Humility, Humor and Curiosity.

Paper

Deep Structured Neural Network with Joint Bottom-up and Top-down Inference
  Wenguan Wang*, Yuanlu Xu*, Quanshi Zhang, Jianbin Shen, Song-Chun Zhu
  under review, 2017. (* contributed equally).
A Causal And-Or Graph Model for Occlusion Fluent Reasoning in Multiple Object Tracking
  Lei Qin*, Yuanlu Xu*, Xiaobai Liu, Song-Chun Zhu
  under review, 2017. (* contributed equally).   [arXiv]
Estimating Human 3D Poses by Hierarchical Grammar Network
  Haoshu Fang*, Yuanlu Xu*, Wenguan Wang, Xiaobai Liu, Song-Chun Zhu
  AAAI Conference on Artificial Intelligence, (AAAI), 2018. (* contributed equally).   [arXiv]
Joint Parsing of Cross-view Scenes with Spatio-temporal Semantic Parse Graph
  Hang Qi*, Yuanlu Xu*, Tao Yuan*, Tianfu Wu, Song-Chun Zhu
  AAAI Conference on Artificial Intelligence, (AAAI), 2018. (* contributed equally).   [arXiv]
Cross-view People Tracking by Scene-centered Spatio-temporal Attributed Graph Parsing
  Yuanlu Xu, Xiaobai Liu, Yang Liu, Song-Chun Zhu
  IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), under review, 2017.
A Stochastic Attribute Grammar for Robust Cross-View Human Tracking
  Xiaobai Liu, Yuanlu Xu, Lei Zhu, Yadong Mu
  IEEE Transactions on Circuits and Systems for Video Technology (TCSVT), under review, 2017.
Place-centric Visual Urban Perception with Hierarchical Deep Multi-instance Regression
  Xiaobai Liu, Qi Chen, Lei Zhu, Yuanlu Xu, Liang Lin
  ACM Multimedia Conference, (ACMMM), 2017.   [pdf]   [BibTex]
Cross-view People Tracking by Scene-centered Spatio-temporal Parsing
  Yuanlu Xu, Xiaobai Liu, Lei Qin, Song-Chun Zhu
  AAAI Conference on Artificial Intelligence (AAAI), 2017.   [pdf]   [BibTex]   [oral]
Multi-view People Tracking via Hierarchical Trajectory Composition
  Yuanlu Xu, Xiaobai Liu, Yang Liu, Song-Chun Zhu
  IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016.   [pdf]   [BibTex]   [poster]   [dataset]
Person Search in a Scene by Jointly Modeling People Commonness and Person Uniqueness
  Yuanlu Xu, Bingpeng Ma, Rui Huang, Liang Lin
  ACM Multimedia Conference (ACMMM), 2014.   [pdf]   [BibTex]   [poster]
Complex Background Subtraction by Pursuing Dynamic Spatio-Temporal Models
  Liang Lin, Yuanlu Xu, Xiaodan Liang, Jian-Huang Lai
  IEEE Transactions on Image Processing (TIP), 23(7): 3191-3202, 2014.   [pdf]   [BibTex]
Detection-free Bayesian Multi-object Tracking via Spatio-Temporal Video Bundles Grouping
  Yongyi Lu, Liang Lin, Yuanlu Xu, Zefeng Lai
  Technical Report, Sun Yat-Sen University, 2013.   [pdf]
Human Re-identification by Matching Compositional Template with Cluster Sampling
  Yuanlu Xu, Liang Lin, Wei-Shi Zheng, Xiaobai Liu
  IEEE International Conference on Computer Vision (ICCV), 2013.   [pdf]   [BibTex]   [poster]   [dataset]
Realtime Object-of-Interest Tracking by Learning Composite Patch-Based Templates
  Yuanlu Xu, Hongfei Zhou, Qing Wang, Liang Lin
  IEEE International Conference on Image Processing (ICIP), 2012.   [pdf]   [BibTex]   [poster]

Thesis

Person Re-identification by Matching Compositional Template with Cluster Sampling
  Yuanlu Xu
  Master Thesis, Sun Yat-Sen University, 2014.   [pdf] (in Chinese)   [defense slides]
Moving Object Segmentation by Pursuing Local Spatio-Temporal Manifolds
  Yuanlu Xu
  Bachelor Thesis, Sun Yat-Sen University, Best Bachelor Thesis Award, 2012.   [pdf] (in Chinese)   [technical report]   [defense slides]

Patent

Person comparison method based on combined template cluster sampling matching
  Patent Publication No.: CN103984916 A, Patent Publication Time: August 13, 2014.   [pdf] (in Chinese)
Moving object segmenting method based on local space-time manifold learning
  Patent Publication No.: CN102750712 B, Patent Publication Time: Jun 17, 2015.   [pdf] (in Chinese)