Stereoscopic view synthesis based on region-wise rendering and sparse representation |
| |
| Affiliation: | 1. Nanyang Normal University, Nanyang, Henan 473061, China;2. Center for Internet of Things, Institute of Microelectronics of Chinese Academy of Sciences, Beijing 100029, China;3. Hebei University of Technology, Tianjin 300401, China;1. Faculty of Engineering, Multimedia University, Persiaran Multimedia, 63100 Cyberjaya, Selangor, Malaysia;2. Faculty of Engineering & Science, Curtin University, 98000 Miri, Sarawak, Malaysia;3. Faculty of Computer Science & Information Technology, University of Malaya, 50603 Kuala Lumpur, Malaysia;1. Telecom Bretagne, Département Image et Traitement de l''Information, Brest 29238, France;2. INSERM U1101, Laboratoire de Traitement de l''Information Médicale, Brest 29238, France;3. MEDECOM, Plougastel Daoulas 29470, France;1. School of Automation Science and Electrical Engineering Beihang University, Beijing, China;2. Computer Science and Digital Technologies Department, Northumbria University, Newcastle, UK;3. Computer Vision Institute, School of Computer Science and Software Engineering, Shenzhen University, Shenzhen, China;4. School of Computer & Software, Nanjing University of Information Science & Technology, China;1. University of Southern California, Los Angeles, CA, USA;2. National Taiwan University, Taipei, Taiwan;1. Universidad Carlos III de Madrid, Leganés, Spain;2. Universidad Politécnica de Madrid, Madrid, Spain |
| |
| Abstract: | Depth image-based rendering (DIBR), which is used to render virtual views with a color image and the corresponding depth map, is one of the key techniques in the 2D to 3D conversion process. One of the main problems in DIBR is how to reduce holes that occur on the generated virtual view images. In this paper, we make two main contributions to deal with the problem. Firstly, a region-wise rendering framework, which divides the original image regions into three special classes and renders each with optimal adaptive process respectively, is introduced. Then, a novel sparse representation-based inpainting method, which can yield visually satisfactory results with less computational complexity for high quality 2D to 3D conversion, is proposed. Numerical experimental results demonstrate the good performance of the proposed methods. |
| |
| Keywords: | Depth image-based rendering Region-wise rendering Sparse background dictionary Stereoscopic image generation |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
