Xinggang Lin

Publications (64)23.62 Total impact

• Article: Stereo Matching Using Local Plane Fitting in Confidence-Based Support Window.

  Chenbo Shi, Guijin Wang, Xiaokang Pei, Bei He, Xinggang Lin
  IEICE Transactions. 01/2012; 95-D:699-702.
• Source

  Available from: mirlab.org

  Conference Proceeding: Multiple instance tracking based on hierarchical maximizing bag's margin boosting

  Chunxiao Liu, Guijin Wang, Xinggang Lin, Bobo Zeng
  [show abstract] [hide abstract]
  ABSTRACT: In online tracking, the tracker evolves to reflect variations in object appearance and surroundings. This updating process is formulated as a supervised learning problem, thus a slight inaccuracy of the tracker will degrade the updating. Multiple Instance Learning (MIL) is used to alleviate such a problem by representing training samples in bags of image patches (or called instances). Difficulties are then passed on to the learning method to train a classifier that discovers the most accurate instance. This paper proposes a Maximizing Bag's Margin (MBM) criteria for MIL. Combined with MBM, a hierarchical boosting is proposed for updating, in which bag and instance weights are introduced to guide classifier retrain ing. Our approach effectively improves the updating's efficiency with less computation cost. Experiments demonstrate the benefits of our method.
  Acoustics, Speech and Signal Processing (ICASSP), 2011 IEEE International Conference on; 06/2011 · 4.63 Impact Factor
• Article: Partial Derivative Guidance for Weak Classifier Mining in Pedestrian Detection.

  Chang Liu, Guijin Wang, Chunxiao Liu, Xinggang Lin
  IEICE Transactions. 01/2011; 94-D:1721-1724.
• Article: Drastic Anomaly Detection in Video Using Motion Direction Statistics.

  Chang Liu, Guijin Wang, Wenxin Ning, Xinggang Lin
  IEICE Transactions. 01/2011; 94-D:1700-1707.
• Article: DSP-Based Parallel Implementation of Speeded-Up Robust Features.

  Chao Liao, Guijin Wang, Quan Miao, Zhiguo Wang, Chenbo Shi, Xinggang Lin
  IEICE Transactions. 01/2011; 94-D:930-933.
• Article: A Framework of Real Time Hand Gesture Vision Based Human-Computer Interaction.

  Liang Sha, Guijin Wang, Xinggang Lin, Kongqiao Wang
  IEICE Transactions. 01/2011; 94-A:979-989.
• Article: A new framework for on-line object tracking based on SURF.

  Quan Miao, Guijin Wang, Chenbo Shi, Xinggang Lin, Zhiwei Ruan
  Pattern Recognition Letters. 01/2011; 32:1564-1571.
• Article: Implementation of Scale and Rotation Invariant On-Line Object Tracking Based on CUDA.

  Quan Miao, Guijin Wang, Xinggang Lin
  IEICE Transactions. 01/2011; 94-D:2549-2552.
• Article: High-Accuracy Sub-Pixel Registration for Noisy Images Based on Phase Correlation.

  Bei He, Guijin Wang, Xinggang Lin, Chenbo Shi, Chunxiao Liu
  IEICE Transactions. 01/2011; 94-D:2541-2544.
• Conference Proceeding: Anomaly detection in surveillance video using motion direction statistics

  Chang Liu, Guijin Wang, Wenxin Ning, Xinggang Lin, Liang Li, Zhou Liu
  [show abstract] [hide abstract]
  ABSTRACT: A novel approach for detecting anomaly in visual surveillance system is proposed in this paper. It is composed of three parts:(a) a dense motion field and motion statistics method, (b) one-class SVM for one-class classification, (c) motion directional PCA for feature dimensionality reduction. Experiments demonstrate the effectiveness of proposed algorithm in detecting abnormal events in surveillance video, while keeping a low false alarm rate. Moreover, it works well in complicated situation where the common tracking or detection module won't work.
  Image Processing (ICIP), 2010 17th IEEE International Conference on; 10/2010
• Conference Proceeding: Topology based affine invariant descriptor for MSERs

  Chenbo Shi, Guijin Wang, Xinggang Lin, Yongming Wang, Chao Liao, Quan Miao
  [show abstract] [hide abstract]
  ABSTRACT: This paper introduces a topology based affine invariant descriptor for maximally stable extremal regions (MSERs). The popular SIFT descriptor computes the texture information on a grey-scale patch. Instead our descriptor use only the topology and geometric information among MSERs so that features can be rapidly matched regardless of the texture in the image patch. Based on the ellipses fitting for the detected MSERs, geometric affine invariants between ellipses pair are extracted as the descriptors. Finally topology based voting selector is designed to achieve the best correspondences. Experiment shows that our descriptor is not only computational faster than SIFT descriptor, but also has better performance on wide angle of view and nonlinear illumination change. In addition, our descriptor shows a good result on multi sensor images registration.
  Image Processing (ICIP), 2010 17th IEEE International Conference on; 10/2010
• Article: Real-Time Human Detection Using Hierarchical HOG Matrices.

  Guan Pang, Guijin Wang, Xinggang Lin
  IEICE Transactions. 01/2010; 93-D:658-661.
• Article: Kernel Based Image Registration Incorporating with Both Feature and Intensity Matching.

  Quan Miao, Guijin Wang, Xinggang Lin
  IEICE Transactions. 01/2010; 93-D:1317-1320.
• Article: Robust Object Tracking via Combining Observation Models.

  Fan Jiang, Guijin Wang, Chang Liu, Xinggang Lin, Weiguo Wu
  IEICE Transactions. 01/2010; 93-D:662-665.
• Article: A Flow-Aware Opportunistic Routing Protocol for Wireless Mesh Networks.

  Haisheng Wu, Guijin Wang, Xinggang Lin
  IEICE Transactions. 01/2010; 93-B:3161-3164.
• Conference Proceeding: Scale and rotation invariant feature-based object tracking via modified on-line boosting.

  Quan Miao, Guijin Wang, Xinggang Lin, Yongming Wang, Chenbo Shi, Chao Liao
  Proceedings of the International Conference on Image Processing, ICIP 2010, September 26-29, Hong Kong, China; 01/2010
• Article: Online HOG Method in Pedestrian Tracking.

  Chang Liu, Guijin Wang, Fan Jiang, Xinggang Lin
  IEICE Transactions. 01/2010; 93-D:1321-1324.
• Article: Measuring Particles in Joint Feature-Spatial Space.

  Liang Sha, Guijin Wang, Anbang Yao, Xinggang Lin
  IEICE Transactions. 01/2009; 92-A:1737-1742.
• Source

  Available from: tsinghua.edu.cn

  Conference Proceeding: Hand posture recognition in video using multiple cues.

  Liang Sha, Guijin Wang, Anbang Yao, Xinggang Lin, Xiujuan Chai
  Proceedings of the 2009 IEEE International Conference on Multimedia and Expo, ICME 2009, June 28 - July 2, 2009, New York City, NY, USA; 01/2009
• Conference Proceeding: Kernel based articulated object tracking with scale adaptation and model update

  Anbang Yao, Guijin Wang, Xinggang Lin, Hao Wang
  [show abstract] [hide abstract]
  ABSTRACT: Kernel based object tracking (KBOT) is one of the most popular and effective techniques for tracking task. However the constancy of the target model and unsound scale adaptation method are two main limitations. In this paper, we present a kernel based approach incorporated with scale estimation and target model update for articulated object tracking task. After predicating the object center with scale fixed KBOT, we extend scale selection theory to estimate the local optimal object scale. Once the object scale has been estimated, a kernel density estimation based strategy is developed to update the target model. Experimental results show that our approach is superior to traditional KBOT in the following two aspects: 1) it is less affected by the object scale change; 2) it is less prone to appearance variation.
  Acoustics, Speech and Signal Processing, 2008. ICASSP 2008. IEEE International Conference on; 05/2008 · 4.63 Impact Factor