키넥트로부터 획득된 깊이 영상의 확장을 위한 GPU 기반 채움 기법
GPU Based Filling Methods for Expansion of Depth Image Captured by Kinect
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Three-dimensional(3D) display technique is widely used in our daily life. Especially, to product augmented game contents which can interact with users, it is necessary to get high quality resolution image data to reconstruct 3D model more exquisitely. In this paper, we tried to expand depth image captured by Kinect using various interpolation methods(nearest neighbor, bilinear, bicubic) to adapt it to the size of original Kinect color image. To measure the quality of expanded depth image compared to original depth image, we used PSNR(Peak Signal-to-noise ratio) index. Besides, we implemented GPU parallel processing algorithm with OpenCL to interpolate a large amount of image data rapidly. As a result of the experiment, a bicubic interpolation method made an accurate depth image although it had a long time.
Abstract
Three-dimensional(3D) display technique is widely used in our daily life. Especially, to product augmented game contents which can interact with users, it is necessary to get high quality resolution image data to reconstruct 3D model more exquisitely. In this paper, we tried to expand depth image captured by Kinect using various interpolation methods(nearest neighbor, bilinear, bicubic) to adapt it to the size of original Kinect color image. To measure the quality of expanded depth image compared to original depth image, we used PSNR(Peak Signal-to-noise ratio) index. Besides, we implemented GPU parallel processing algorithm with OpenCL to interpolate a large amount of image data rapidly. As a result of the experiment, a bicubic interpolation method made an accurate depth image although it had a long time.
참고문헌 (13)
- R. S. Wagner, "Image super resolution for improved automatic target recognition", Proc. SPIE, 5426 (2004)
- W. Freeman, "Example based super resolution", IEEE Computer Graphics and Applications, 22(2) (2002)
- H. Chang, "Super resolution through neighbor embedding", Proc. IEEE CVPR, 1 (2004)
- 박영우, "키넥트 2.0을 활용한 게임 모션 라이브러리 툴 구축", 한국컴퓨터게임학회논문지, 28(2), pp.79~85 (2015), (사)한국컴퓨터게임학회
- 권순철, "키넥트 깊이 정보와 DSLR을 이용한 고해상도 3D 객체 생성", 한국컴퓨터게임학회논문지, 26(1), pp.221~227 (2013), (사)한국컴퓨터게임학회
- 박태정, "게임 오브젝트 Polygon Soup 모델의 근사화 및 보간을 위한 고속 병렬 GPU 적분형 MLS 연산", 한국컴퓨터게임학회논문지, 25(1), pp.179~192 (2012), (사)한국컴퓨터게임학회[DOI]
- Ji-Seong Jeong, "Development of a real-time integral imaging display system based on graphics processing unit parallel processing using a depth camera", Optical Engineering, 53(1) (2014)
- Aaftab Munshi, "The OpenCL Specification, Version: 1.1", Khronos OpenCL Working Group, ~ (2010)
- NVIDIA, ""CUDA C programming guide" Ver. 3.3.3", ~ (2010)
- NVIDIA, ""OpenCL programming guide for the CUDA architecture" Ver. 5.5", ~ (2013)
- Sun-Gyu Hwang, "Image Processing Programming by Visual C++" (2007), Hanbit media
- "Peak signal-to-noise ratio" (2016)[URL]
- "Kinect for Windows SDK 2.0"[URL]