Title :
2D to 3D video evaluation
Author :
Li Yueli ; Huicheng Lian
Author_Institution :
Sch. of Comput. Eng. & Sci., Shanghai Univ., Shanghai, China
Abstract :
A novel and simple method is proposed to predict the video results of 2D to 3D in this paper. Firstly, the motion features of 2D video frames for converting to 3D are extracted. Then, the motion feature difference between the adjacent two frames is computed and a threshold operation for denoising is adopted. Thirdly, some frames which meet certain conditions were chosen to compute the mean motion difference. Finally, the prediction score of 2D to 3D video results is obtained based on the computed mean motion difference. The experimental results indicate that the proposed approach can perform well for predicting the video results of 2D to 3D, especially for cartoon.
Keywords :
feature extraction; image denoising; image motion analysis; image segmentation; prediction theory; statistical analysis; video signal processing; 2D video evaluation; 2D video frames; 3D video evaluation; cartoon; mean motion difference; motion feature difference; motion features extraction; prediction score; threshold operation; video denoising; 2D to 3D; motion feature difference; threshold operation;
Conference_Titel :
Information Science and Control Engineering 2012 (ICISCE 2012), IET International Conference on
Conference_Location :
Shenzhen
Electronic_ISBN :
978-1-84919-641-3
DOI :
10.1049/cp.2012.2325