CVPG @ NWPU
中文 | English

Intro

CVPG (Computer Vision and Computational Photography Group) at  Northwestern Polytechnical University (CVPG@NPU), founded in the year of 2002, is affiliated with Shaanxi Provincial Key Laboratory of Speech and Image Information Processing (SAIIP). Our researh topics cover fundamental theories and technologies in 3D reconstruction, light field imaging and processing, computational photography and application, object detection and tracking in video etc. In the year of 2018, the group consists of one full professor, two associate professors, one lecture, four adjunct professors and over twenty PhD and master students. We have established extensive collaborations with famous universities in USA, such as CMU, Stanford, MIT, UC Berkeley, and well known universities and reseach institutes in China, such as Tsinghua, Beihang, NUDT, Sichuan, Academy of Opto-Electronics, Xi'an Institute of Optics and Precision Mechanics, Shanghai Institute of Optics and Fine Mechanics, Institute of Optics and Electronics (Chengdu), CAS. In recent years, we have taken more than twenty research projects supported by NSFC, MOST, MOE, Shaanxi Province and industrial partners.

Topic

Light Field Imaging and Processing

Light-field technology heralds one of the biggest changes to imaging since 1826, when Joseph-Nicéphore Niépce made the first permanent photograph of a scene from nature. A single light-field snapshot can provide photos where focus, exposure, and even depth of field are adjustable after the picture is taken. Light-field imaging is far more ambitious. Instead of merely recording the sum of all the light rays falling on each photosite, a light-field camera aims to measure the intensity and direction of every incoming ray. With that information, you can generate not just one but every possible image of whatever is within the camera’s field of view at that moment. The information a light-field camera records is, mathematically speaking, part of something that optics specialists call the plenoptic function. This function describes the totality of light rays filling a given region of space at any one moment. It’s a function of five dimensions, because you need three (x, y, and z) to specify the position of each vantage point, plus two more (often denoted θ and φ) for the angle of every incoming ray.


Selected Publications

4D Light Field Segmentation from Light Field Super-pixel Hypergraph Representation [PDF] [BIB]

Xianqiang Lv, Xue Wang, Qing Wang, Jingyi Yu
IEEE Transactions on Visualization and Computer Graphics, 2020, Early Access (2020-03)

Revisiting Spatio-Angular Trade-off in Light Field Cameras and Extended Applications in Super-Resolution [PDF] [BIB]

Hao Zhu, Mantang Guo, Hongdong Li, Qing Wang, Antonio Robles-Kelly
IEEE Transactions on Visualization and Computer Graphics, 2020 (2019-12)

Full View Optical Flow Estimation Leveraged from Light Field Superpixel [PDF] [BIB]

Hao Zhu, Xiaoming Sun, Qi Zhang, Qing Wang, Antonio Robles-Kelly, Hongdong Li, Shaodi You
IEEE Transactions on Computational Imaging, 6(1):12-23, 2020. (2020-01)

4D Light Field Superpixel and Segmentation [PDF] [BIB]

Hao Zhu, Qi Zhang, Qing Wang, Hongdong Li
IEEE Transactions on Image Processing, 29(1):85-99, 2020. (2020-01)

Ray-Space Projection Model for Light Field Camera [PDF] [BIB]

Qi Zhang, Jinbo Ling, Qing Wang, Jingyi Yu
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, USA, pp. 10121-10129 (2019-06)

A Generic Multi-Projection-Center Model and Calibration Method for Light Field Cameras [PDF] [BIB]

Qi Zhang, Chunping Zhang, Jinbo Ling, Qing Wang, Jingyi Yu
IEEE Transactions on Pattern Analysis and Machine Intelligence, 41(11):2539-2552, 2019. (2019-11)

Motion-based Temporal Alignment of Independently Moving Cameras [PDF] [BIB]

Xue Wang, Jianbo Shi, Hyun Soo Park, Qing Wang
IEEE Transactions on Circuits and Systems for Video Technology, 27(11):2344-2354, 2017 (2017-11)

High Angular Resolution Light Field Reconstruction with Coded-Aperture Mask [PDF] [BIB]

Wanxin Qu, Guoqing Zhou, Hao Zhu, Zhaolin Xiao, Qing Wang, Rene Vidal
IEEE Int. Conf. on Image Processing (ICIP), Beijing, China, pp.3036-3040 (2017-09)

Extending the FOV from Disparity and Color Consistencies in Multiview Light Fields [PDF] [BIB]

Zhao Ren, Qi Zhang, Hao Zhu, Qing Wang
IEEE Int. Conf. on Image Processing (ICIP), Beijing, China, pp.1157-1161 (2017-09)

Seeing Beyond Foreground Occlusion: A Joint Framework for SAP Based Scene Depth and Appearance Reconstruction [PDF] [BIB]

Zhaolin Xiao, Lipeng Si, Guoqing Zhou
IEEE Journal of Selected Topics in Signal Processing, 11(7):979-991, 2017 (2017-10)

Light Field Imaging: Models, Calibrations, Reconstructions, and Applications [PDF] [BIB]

Hao Zhu, Qing Wang, Jingyi Yu
Frontiers of Information Technology & Electronic Engineering, 18(9):1236-1249, 2017 (2017-09)

Occlusion-Model Guided Anti-Occlusion Depth Estimation in Light Field [PDF] [BIB]

Hao Zhu, Qing Wang, Jingyi Yu
IEEE Journal of Selected Topics in Signal Processing, 11(7):965-978, 2017 (2017-10)

4D Light Field Superpixel and Segmentation [PDF] [BIB]

Hao Zhu, Qi Zhang, Qing Wang
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Hawaii, USA, pp. 6709-6717 (2017-07)

Aliasing Detection and Reduction Scheme on Angularly Undersampled Light Fields [PDF] [BIB]

Zhaolin Xiao, Qing Wang, Guoqing Zhou, Jingyi Yu
IEEE Transactions on Image Processing, 26(5): 2103-2115, 2017 (2017-05)

Dense Depth-map Estimation and Geometry Inference from Light Fields via Global Optimization [PDF] [BIB]

Lipeng Si, Qing Wang
Asian Conference on Computer Vision (ACCV), Taipei, China, pp.83-98, 2017 (2016-11)

Survey on Imaging Model and Calibration of Light Field Camera [PDF] [BIB]

Chunping Zhang, Qing Wang
中国激光(Chinese Journal of Lasers), 43(6):0609004, 2016 (2016-06)

Depth Estimation from Light Field Analysis Based Multiple Cues Fusion [PDF] [BIB]

Degang Yang, Zhaolin Xiao, Heng Yang, Qing Wang
计算机学报 (Chinese Journal of Computers), 38(12):2437-2449 (2015-12)

Aliasing Detection and Reduction in Plenoptic Imaging [PDF] [BIB]

Zhaolin Xiao, Qing Wang, Guoqing Zhou, Jingyi Yu
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Columbus, USA, pp. 3326-3333 (2014-06)

Comuptational Photography and Application

Computational photography or computational imaging refers to digital image capture and processing techniques that use digital computation instead of optical processes. Computational photography can improve the capabilities of a camera, or introduce features that were not possible at all with film based photography, or reduce the cost or reduce the size of camera elements. Examples of computational photography include in-camera computation of digital panoramas, high-dynamic-range images, and light field cameras. Light field cameras use novel optical elements to capture three dimensional scene information which can then be used to produce 3D images, enhanced of depth-of-field, and selective de-focusing (or "post focus"). Enhanced depth-of-field reduces the need for mechanical focusing systems. All of these features use computational imaging techniques.

Selected Publications

A video-driven method to generate water surface models [PDF] [BIB]

Cong Chen, Qing Wang
计算机应用研究 (2014-03)

An Improved Triangle Star Identification Method Based on Multi-FOV Star Sensor [PDF] [BIB]

Xue Xiong, Qing Wang
计算机测量与控制 (2014-01)

A Multi-scale Anti-aliasing Rendering Algorithm for the Light Field Imaging [PDF] [BIB]

Zhaolin Xiao, Qing Wang, Guoqing Zhou, Heng Yang
计算机辅助设计与图形学学报 (Journal of CAD and CG), 26(7):1126-1134 (2014-07)

e-Silkroad: A Sample of Combining Social Media with Cultural Tourism [PDF] [BIB]

Qing Wang, Xiaozhen Qi, Jiong Xu
ACM MM 2010 Workshop on Connected Multimedia, 27-32 (2010-10)

Modeling Urban Scenes in the Spatial-Temporal Space [PDF] [BIB]

Jiong Xu, Qing Wang, Jie Yang
ACCV 2010, 374-387 (2011-11)

3D Reconstruction and Representation

3D reconstruction is one of fundmental tasks in computer vision. One can reconstruct geometric structure of real wold scene by estimating the depth and poses of cameras, which provides infrastructural data for object detection, segmentation and background modeling.

Selected Publications

Accurate 3D Reconstruction from Circular Light Field Using CNN-LSTM [PDF] [BIB]

Zhengxi Song, Hao Zhu, Qi Wu, Xue Wang, Hongdong Li, Qing Wang
IEEE Int. Conf. on Multimedia and Expo (ICME), London, UK, 2020 (2020-05)

A Generic Multi-Projection-Center Model and Calibration Method for Light Field Cameras [PDF] [BIB]

Qi Zhang, Chunping Zhang, Jinbo Ling, Qing Wang, Jingyi Yu
IEEE Transactions on Pattern Analysis and Machine Intelligence, 41(11):2539-2552, 2019. (2019-11)

Motion-based Temporal Alignment of Independently Moving Cameras [PDF] [BIB]

Xue Wang, Jianbo Shi, Hyun Soo Park, Qing Wang
IEEE Transactions on Circuits and Systems for Video Technology, 27(11):2344-2354, 2017 (2017-11)

Seeing Beyond Foreground Occlusion: A Joint Framework for SAP Based Scene Depth and Appearance Reconstruction [PDF] [BIB]

Zhaolin Xiao, Lipeng Si, Guoqing Zhou
IEEE Journal of Selected Topics in Signal Processing, 11(7):979-991, 2017 (2017-10)

Robust Outlier Removal Using Penalized Linear Regression in Multiview Geometry [PDF] [BIB]

Guoqing Zhou, Qing Wang, Zhaolin Xiao
Neurocomputing, 267:455-465, 2017 (2017-08)

Aliasing Detection and Reduction Scheme on Angularly Undersampled Light Fields [PDF] [BIB]

Zhaolin Xiao, Qing Wang, Guoqing Zhou, Jingyi Yu
IEEE Transactions on Image Processing, 26(5): 2103-2115, 2017 (2017-05)

Video Synchronization with Trajectory Pulse [PDF] [BIB]

Xue Wang, Qing Wang
4th Chinese Conference on Intelligent Visual Surveillance (2016-10)

Sparse Representation with Geometric Configuration Constraint for Line Segment Matching [PDF] [BIB]

Qing Wang, Tingwang Chen, Lipeng Si
Neurocomputing, 134(25):100-110, 2014 (2014-01)

Reconstructing Scene Depth and Appearance behind Foreground Occlusion Using Camera Array [PDF] [BIB]

Zhaolin Xiao, Qing Wang, Lipeng Si, Guoqing Zhou
ICIP 2014, 41-45 (2014-10)

A Resection Method Based on Enhanced Continuous Taboo Search [PDF] [BIB]

Guoqing Zhou, Qing Wang
电子学报 (ACTA ELECTRONICA SINICA), 42(12):2422-2428 (2014-12)

Enhanced Continuous Tabu Search for Parameter Estimation in Multiview Geometry [PDF] [BIB]

Guoqing Zhou, Qing Wang
ICCV 2013, 3240-3247 (2013-12)

A Triangulation Method Based on minmaxKKT [PDF] [BIB]

Guoqing Zhou, Qing Wang
自动化学报 (ACTA AUTOMATICA SINICA), 38(9):1439-1444 (2012-09)

Sparse Representation with Geometric Configuration Constraint for Line Segment Matching [PDF] [BIB]

Qing Wang, Tingwang Chen
IScIDE 2011, LNCS 7202:498-505 (2011-10)

Modeling Urban Scenes in the Spatial-Temporal Space [PDF] [BIB]

Jiong Xu, Qing Wang, Jie Yang
ACCV 2010, 374-387 (2011-11)

3D Line Segment Detection for Unorganized Point Clouds from Multi-view Stereo [PDF] [BIB]

Tingwang Chen, Qing Wang
ACCV 2010, 400-411 (2010-11)

Distance-Based Multiple Paths Quantization of Vocabulary Tree for Object and Scene Retrieval [PDF] [BIB]

Heng Yang, Qing Wang, Ellen Yi-Luen Do
ACCV 2009, 313-322 (2010-01)

Robust Wide Baseline Point Matching Based on Scale Invariant Feature Descriptor [PDF] [BIB]

Sicong Yue, Qing Wang, Rongchun Zhao
Chinese Journal of Aeronautics, 22(1):70-74 (2009-02)

Multiple Unordered Wide-Baseline Image Matching and Grouping [PDF] [BIB]

Zhoucan He, Qing Wang, Heng Yang
ICME 2009, 690-693 (2009-07)

Distance-based Multiple Paths Quantization of Vocabulary Tree for Object and Scene Retrieval [PDF] [BIB]

Heng Yang, Qing Wang, Ellen Do
ACCV 2009, 313-322 (2009-09)

Grouping and Organizing Unordered Images for Multi-View Feature Correspondences [PDF] [BIB]

Zhoucan He, Qing Wang
ICIG 2009, 490-495 (2009-09)

Efficient Scene Image Clustering for Internet Collections [PDF] [BIB]

Heng Yang, Qing Wang, Zhoucan He
ICIG 2009, 471-476 (2009-09)

A Novel Local Feature Descriptor for Image Matching [PDF] [BIB]

Heng Yang, Qing Wang
ICME 2008, 1405-1408 (2008-06)

Indexing Sub-Vector Distance for High-Dimensional Feature Matching [PDF] [BIB]

Heng Yang, Qing Wang, Zhoucan He
BMVC 2008 (2008-09)

Randomized Sub-vectors Hashing for High-dimensional Image Feature Matching [PDF] [BIB]

Heng Yang, Qing Wang, Zhoucan He
ACM Multimedia 2008, 705-708 (2008-10)

Object Detection and Tracking

Object Detection and Tracking is one of important branches in reseach domain of computer vision.

Selected Publications

Figure-Aware Tracking under Occlusion from Monocular Videos [PDF] [BIB]

Xue Wang, Qing Wang
ICVRV 2014, 116-121 (2014-08)

Coupled Data Association and L2 Minimization for Multiple Object Tracking under Occlusion [PDF] [BIB]

Xue Wang, Qing Wang
SPIE/COS Photonics Asia, 9273 (2014-11)

Tour Routes Recommendation Begins with Multimodal Classification [PDF] [BIB]

Xiujun Chen, Qing Wang
Journal of Multimedia, 7(1):21-30 (2012-01)

An Image Classification Approach Based on Sparse Coding and Multiple Kernel Learning [PDF] [BIB]

Xiaozhen Qi, Qing Wang
电子学报 (ACTA ELECTRONICA SINICA), 40(4):773-779 (2012-04)

Spatio-Temporal Clustering Model for Multi- Object Tracking through Occlusions [PDF] [BIB]

Lei Zhang, Qing Wang
ACCV 2012, 177-190 (2012-11)

A Comprehensive Evaluation on Non-deterministic Motion Estimation [PDF] [BIB]

Changzhu Wu, Qing Wang
ICPR 2010, 2333-2336 (2010-08)

Drinking Activity Analysis from Fast Food Eating Video Using Generative Models [PDF] [BIB]

Qing Wang, Jie Yang
ACM MM 2009 Workshop on Multimedia for Cooking and Eating activities, 31-38 (2009-10)

Motion Estimation Approach Based on Dual-tree Complex Wavelets [PDF] [BIB]

Changzhu Wu, Qing Wang, Hongxiao Wang
ICPR 2008, 1-4 (2008-12)

A Novel Human Gait Recognition Method by Segmenting and Extracting the Region Variance Feature [PDF] [BIB]

Yanmei Chai, Qing Wang, Jingping Jia, Rongchun Zhao
ICPR 2006, 425-428 (2006-08)

QP_TR: Trust Region Blob Tracking Through Scale-Space [PDF] [BIB]

Jingping Jia, Qing Wang, Yanmei Chai, Rongchun Zhao
ICIP 2006, 1781-1784 (2006-08)

QP_TR: Trust Region Blob Tracking Through Scale-Space with Automatic Selection of Features [PDF] [BIB]

Jingping Jia, Qing Wang, Yanmei Chai, Rongchun Zhao
ICIAR (2006-09)