publications: - title: "Online Real Boosting for Object Tracking Under Severe Appearance Changes and Occlusion" author: - name: "Li Xu" link: "http://appsrv.cse.cuhk.edu.hk/~xuli/" - name: "Takayoshi Yamashita" link: "http://vision.cs.chubu.ac.jp/~takayoshi" - name: "Shihong Lao" link: "https://researchr.org/alias/shihong-lao" - name: "Masato Kawade" link: "https://researchr.org/alias/masato-kawade" - name: "Feihu Qi" link: "https://researchr.org/alias/feihu-qi" year: "2007" month: "April" doi: "http://dx.doi.org/10.1109/ICASSP.2007.366060" abstract: "Robust visual tracking is always a challenging but yet intriguing problem owing to the appearance variability of target objects. In this paper we propose a novel method to handle large changes in appearance based on online real-value boosting, which is utilized to incrementally learn a strong classifier to distinguish between objects and their background. By incorporating online real boosting into a particle filter framework, our tracking algorithm shows a strong adaptability for different target objects which undergo severe appearance changes during the tracking process." links: doi: "http://dx.doi.org/10.1109/ICASSP.2007.366060" tags: - "rule-based" - "meta-model" - "Meta-Environment" - "incremental" - "meta-objects" researchr: "https://researchr.org/publication/XuYLKQ07" cites: 0 citedby: 0 booktitle: "ICASSP" kind: "inproceedings" key: "XuYLKQ07" - title: "A Segmentation Based Variational Model for Accurate Optical Flow Estimation" author: - name: "Li Xu" link: "http://appsrv.cse.cuhk.edu.hk/~xuli/" - name: "Jianing Chen" link: "https://researchr.org/alias/jianing-chen" - name: "Jiaya Jia" link: "https://researchr.org/alias/jiaya-jia" year: "2008" doi: "http://dx.doi.org/10.1007/978-3-540-88682-2_51" abstract: "Segmentation has gained in popularity in stereo matching. However, it is not trivial to incorporate it in optical flow estimation due to the possible non-rigid motion problem. In this paper, we describe a new optical flow scheme containing three phases. First, we partition the input images and integrate the segmentation information into a variational model where each of the segments is constrained by an affine motion. Then the errors brought in by segmentation are measured and stored in a confidence map. The final flow estimation is achieved through a global optimization phase that minimizes an energy function incorporating the confidence map. Extensive experiments show that the proposed method not only produces quantitatively accurate optical flow estimates but also preserves sharp motion boundaries, which makes the optical flow result usable in a number of computer vision applications, such as image/video segmentation and editing." links: doi: "http://dx.doi.org/10.1007/978-3-540-88682-2_51" tags: - "optimization" - "rule-based" - "meta-model" - "data-flow" - "information models" - "Meta-Environment" - "partitioning" researchr: "https://researchr.org/publication/XuCJ08" cites: 0 citedby: 0 pages: "671-684" booktitle: "eccv" kind: "inproceedings" key: "XuCJ08" - title: "Stereo Matching: An Outlier Confidence Approach" author: - name: "Li Xu" link: "http://appsrv.cse.cuhk.edu.hk/~xuli/" - name: "Jiaya Jia" link: "https://researchr.org/alias/jiaya-jia" year: "2008" doi: "http://dx.doi.org/10.1007/978-3-540-88693-8_57" abstract: "One of the major challenges in stereo matching is to handle partial occlusions. In this paper, we introduce the Outlier Confidence (OC) which dynamically measures how likely one pixel is occluded. Then the occlusion information is softly incorporated into our model. A global optimization is applied to robustly estimating the disparities for both the occluded and non-occluded pixels. Compared to color segmentation with plane fitting which globally partitions the image, our OC model locally infers the possible disparity values for the outlier pixels using a reliable color sample refinement scheme. Experiments on the Middlebury dataset show that the proposed two-frame stereo matching method performs satisfactorily on the stereo images." links: doi: "http://dx.doi.org/10.1007/978-3-540-88693-8_57" tags: - "optimization" - "meta-model" - "refinement" - "information models" - "Meta-Environment" - "partitioning" - "systematic-approach" researchr: "https://researchr.org/publication/XuJ08%3A2" cites: 0 citedby: 0 pages: "775-787" booktitle: "eccv" kind: "inproceedings" key: "XuJ08:2" - title: "Detecting and Segmenting Text from Natural Scenes with 2-Stage Classification" author: - name: " Renjie Jiang" link: "https://researchr.org/alias/renjie-jiang" - name: "Feihu Qi" link: "https://researchr.org/alias/feihu-qi" - name: "Li Xu" link: "http://appsrv.cse.cuhk.edu.hk/~xuli/" - name: "Guorong Wu" link: "https://researchr.org/alias/guorong-wu" year: "2006" doi: "http://doi.ieeecomputersociety.org/10.1109/ISDA.2006.253718" abstract: "This paper proposes a novel learning-based approach for detecting and segmenting text from scene images. First, the input image is decomposed into a list of connected-components (CCs) by color clustering algorithm. Then all the CCs including text CCs and non-text CCs are verified by a 2-stage classification module, where most of non-text CCs are discarded by cascade classifier and the remaining CCs are further verified by SVM. All the accepted CCs are output to generate result image. Experiments have been taken on a lot of images with different nature scenes and show satisfactory performance of our proposed method." links: doi: "http://doi.ieeecomputersociety.org/10.1109/ISDA.2006.253718" tags: - "rule-based" - "classification" - "systematic-approach" researchr: "https://researchr.org/publication/Jiang-isda-2006" cites: 0 citedby: 0 booktitle: "isda" kind: "inproceedings" key: "Jiang-isda-2006" - title: "Shadow Removal from a Single Image" author: - name: "Li Xu" link: "http://appsrv.cse.cuhk.edu.hk/~xuli/" - name: "Feihu Qi" link: "https://researchr.org/alias/feihu-qi" - name: " Renjie Jiang" link: "https://researchr.org/alias/renjie-jiang" year: "2006" doi: "http://doi.ieeecomputersociety.org/10.1109/ISDA.2006.253756" abstract: "Shadow detection and removal in real scene images is always a challenging but yet intriguing problem. In contrast with the rapidly expanding and continuous interests on this area, it is always hard to provide a robust system to eliminate shadows in static images. This paper aimed to give a comprehensive method to remove both vague and hard shadows from a single image. First, classification is applied to the derivatives of the input image to separate the vague shadows. Then, color invariant is exploited to distinguish the hard shadow edges from the material edges. Next, we derive the illumination image via solving the standard Poisson equation. Finally, we got the shadow-free reflectance image. Experimental results showed that our method can robustly remove both vague and hard shadows appearing in the real scene images." links: doi: "http://doi.ieeecomputersociety.org/10.1109/ISDA.2006.253756" tags: - "classification" researchr: "https://researchr.org/publication/XuQJ06" cites: 0 citedby: 0 booktitle: "isda" kind: "inproceedings" key: "XuQJ06"