论文信息 - Video Image Mosaics in Real-Time Based on SIFT

Video Image Mosaics in Real-Time Based on SIFT

This paper proposes the method of video image mosaics in real-time based on Scale Invariant Feature Transform (SIFT) algorithm. The real-time processing is great significant for the video image mosaics. SIFT is the efficient method for extracting distinctive invariant features from images. But it costs so much computation time that it can’t meet the real-time processing demand. Therefore, an improved SIFT approach is proposed in this paper to accelerate its execution. The result shows that the method is validated and meets the real-time processing computation demand.

Yang Ping | Qi Feng | Mao Zheng | Gao Anjie

[1] G LoweDavid,et al. Distinctive Image Features from Scale-Invariant Keypoints , 2004 .

[2] Robert C. Bolles,et al. Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography , 1981, CACM.

[3] J. Koenderink. The structure of images , 2004, Biological Cybernetics.

[4] P. Anandan,et al. Mosaic based representations of video sequences and their applications , 1995, Proceedings of IEEE International Conference on Computer Vision.

[5] Thomas Wiegand,et al. SIFT Implementation and Optimization for General-Purpose GPU , 2007 .

[6] M. Meribout. Video Segmentation for Content-based Coding , 2004 .

[7] Yihong Gong,et al. A robust image mosaicing technique capable of creating integrated panoramas , 1999, 1999 IEEE International Conference on Information Visualization (Cat. No. PR00210).