Hansung Kim | University of Southampton (original) (raw)
Papers by Hansung Kim
IEEE Transactions on Multimedia
2019 IEEE Conference on Virtual Reality and 3D User Interfaces (VR)
2020 25th International Conference on Pattern Recognition (ICPR)
European Conference on Visual Media Production
International Journal of Computer Vision
Existing techniques for dynamic scene reconstruction from multiple wide-baseline cameras primaril... more Existing techniques for dynamic scene reconstruction from multiple wide-baseline cameras primarily focus on reconstruction in controlled environments, with fixed calibrated cameras and strong prior constraints. This paper introduces a general approach to obtain a 4D representation of complex dynamic scenes from multi-view wide-baseline static or moving cameras without prior knowledge of the scene structure, appearance, or illumination. Contributions of the work are: an automatic method for initial coarse reconstruction to initialize joint estimation; sparse-to-dense temporal correspondence integrated with joint multi-view segmentation and reconstruction to introduce temporal coherence; and a general robust approach for joint segmentation refinement and dense reconstruction of dynamic scenes by introducing shape constraint. Comparison with state-of-the-art approaches on a variety of complex indoor and outdoor scenes, demonstrates improved accuracy in both multi-view segmentation and ...
2019 IEEE Conference on Virtual Reality and 3D User Interfaces (VR), Mar 1, 2019
ELCVIA Electronic Letters on Computer Vision and Image Analysis
In this paper, we propose a new 3D modeling and rendering system using multiple stereo cameras. W... more In this paper, we propose a new 3D modeling and rendering system using multiple stereo cameras. When target objects are captured by cameras, each capturing PC segments the objects and estimates disparity fields, then they transmit the segmented masks, disparity fields, and color textures of objects to a 3D modeling server. The modeling server generates 3D models of the objects from the gathered masks and disparity fields. Finally, the server generates a video at the designated point of view with the 3D model and texture information from cameras.
2014 IEEE International Conference on Image Processing (ICIP), 2014
We propose a robust 3D feature description and registration method for 3D models reconstructed fr... more We propose a robust 3D feature description and registration method for 3D models reconstructed from various sensor devices. General 3D feature detectors and descriptors generally show low distinctiveness and repeatability for matching between different data modalities due to differences in noise and errors in geometry. The proposed method considers not only local 3D points but also neighbouring 3D keypoints to improve keypoint matching. The proposed method is tested on various multi-modal datasets including LIDAR scans, multiple photos, spherical images and RGBD videos to evaluate the performance against existing methods.
Optical Engineering, 2004
Three-dimensional Imaging, Visualization, and Display, 2009
ABSTRACT One of the most desirable ways of realizing high quality information and telecommunicati... more ABSTRACT One of the most desirable ways of realizing high quality information and telecommunication services has been called "The Sensation of Reality," which can be achieved by visual communication based on 3-D (Three-dimensional) images. These kinds of 3-D imaging systems have revealed potential applications in the fields of education, entertainment, medical surgery, video conferencing, etc. Especially, three-dimensional television (3-D TV) is believed to be the next generation of TV technology. Figure 13.1 shows how TV's display technologies have evolved , and Fig. 13.2 details the evolution of TV broadcasting as forecasted by the ETRI (Electronics and Telecommunications Research Institute). It is clear that 3-D TV broadcasting will be the next development in this field, and realistic broadcasting will soon follow.
Third International Symposium on 3D Data Processing, Visualization, and Transmission (3DPVT'06), 2006
SPIE Proceedings, 2007
In this paper, we propose a new free-view video system that generates 3D video from arbitrary poi... more In this paper, we propose a new free-view video system that generates 3D video from arbitrary point of view, using multiple cameras. When target objects are captured by these cameras, the PC allocated to each capturing camera segments the objects and transmits the masks and color textures to a 3D modeling server via the system's network. The modeling server then
Lecture Notes in Computer Science, 2008
17th International Conference on Artificial Reality and Telexistence (ICAT 2007), 2007
ACM SIGGRAPH 2005 Posters on - SIGGRAPH '05, 2005
ABSTRACT This research paper describes a new method in which the discovery and construction of ma... more ABSTRACT This research paper describes a new method in which the discovery and construction of man-made structures, specifically medieval castles, in a three dimensional environment can be intelligently constructed given only a random terrain and simple user-defined ...
IEEE Transactions on Multimedia
2019 IEEE Conference on Virtual Reality and 3D User Interfaces (VR)
2020 25th International Conference on Pattern Recognition (ICPR)
European Conference on Visual Media Production
International Journal of Computer Vision
Existing techniques for dynamic scene reconstruction from multiple wide-baseline cameras primaril... more Existing techniques for dynamic scene reconstruction from multiple wide-baseline cameras primarily focus on reconstruction in controlled environments, with fixed calibrated cameras and strong prior constraints. This paper introduces a general approach to obtain a 4D representation of complex dynamic scenes from multi-view wide-baseline static or moving cameras without prior knowledge of the scene structure, appearance, or illumination. Contributions of the work are: an automatic method for initial coarse reconstruction to initialize joint estimation; sparse-to-dense temporal correspondence integrated with joint multi-view segmentation and reconstruction to introduce temporal coherence; and a general robust approach for joint segmentation refinement and dense reconstruction of dynamic scenes by introducing shape constraint. Comparison with state-of-the-art approaches on a variety of complex indoor and outdoor scenes, demonstrates improved accuracy in both multi-view segmentation and ...
2019 IEEE Conference on Virtual Reality and 3D User Interfaces (VR), Mar 1, 2019
ELCVIA Electronic Letters on Computer Vision and Image Analysis
In this paper, we propose a new 3D modeling and rendering system using multiple stereo cameras. W... more In this paper, we propose a new 3D modeling and rendering system using multiple stereo cameras. When target objects are captured by cameras, each capturing PC segments the objects and estimates disparity fields, then they transmit the segmented masks, disparity fields, and color textures of objects to a 3D modeling server. The modeling server generates 3D models of the objects from the gathered masks and disparity fields. Finally, the server generates a video at the designated point of view with the 3D model and texture information from cameras.
2014 IEEE International Conference on Image Processing (ICIP), 2014
We propose a robust 3D feature description and registration method for 3D models reconstructed fr... more We propose a robust 3D feature description and registration method for 3D models reconstructed from various sensor devices. General 3D feature detectors and descriptors generally show low distinctiveness and repeatability for matching between different data modalities due to differences in noise and errors in geometry. The proposed method considers not only local 3D points but also neighbouring 3D keypoints to improve keypoint matching. The proposed method is tested on various multi-modal datasets including LIDAR scans, multiple photos, spherical images and RGBD videos to evaluate the performance against existing methods.
Optical Engineering, 2004
Three-dimensional Imaging, Visualization, and Display, 2009
ABSTRACT One of the most desirable ways of realizing high quality information and telecommunicati... more ABSTRACT One of the most desirable ways of realizing high quality information and telecommunication services has been called "The Sensation of Reality," which can be achieved by visual communication based on 3-D (Three-dimensional) images. These kinds of 3-D imaging systems have revealed potential applications in the fields of education, entertainment, medical surgery, video conferencing, etc. Especially, three-dimensional television (3-D TV) is believed to be the next generation of TV technology. Figure 13.1 shows how TV's display technologies have evolved , and Fig. 13.2 details the evolution of TV broadcasting as forecasted by the ETRI (Electronics and Telecommunications Research Institute). It is clear that 3-D TV broadcasting will be the next development in this field, and realistic broadcasting will soon follow.
Third International Symposium on 3D Data Processing, Visualization, and Transmission (3DPVT'06), 2006
SPIE Proceedings, 2007
In this paper, we propose a new free-view video system that generates 3D video from arbitrary poi... more In this paper, we propose a new free-view video system that generates 3D video from arbitrary point of view, using multiple cameras. When target objects are captured by these cameras, the PC allocated to each capturing camera segments the objects and transmits the masks and color textures to a 3D modeling server via the system's network. The modeling server then
Lecture Notes in Computer Science, 2008
17th International Conference on Artificial Reality and Telexistence (ICAT 2007), 2007
ACM SIGGRAPH 2005 Posters on - SIGGRAPH '05, 2005
ABSTRACT This research paper describes a new method in which the discovery and construction of ma... more ABSTRACT This research paper describes a new method in which the discovery and construction of man-made structures, specifically medieval castles, in a three dimensional environment can be intelligently constructed given only a random terrain and simple user-defined ...