摘要: This paper proposes a new technique to build background memory based on mosaicking. More precisely, the first identifies and foreground regions local motion estimates. Camera is then estimated by applying parametric global estimation. Finally, after compensating for camera motion, content temporally integrated in long-term memory. The method leads high coding performances allows content-based functionalities.
sci-hub.se 参考文章(8)
F. Moscheni, F. Dufaux, M. Kunt, A new two-stage global/local motion estimation based on a background/foreground segmentation international conference on acoustics, speech, and signal processing. ,vol. 4, pp. 2261- 2264 ,(1995) , 10.1109/ICASSP.1995.479941
Fabrice Moscheni, Frederic Dufaux, Regions merging based on robust statistical testing visual communications and image processing. ,vol. 2727, pp. 1118- 1129 ,(1996) , 10.1117/12.233184
Frédéric Dufaux, Fabrice Moscheni, Segmentation-Based Motion Estimation for Second Generation Video Coding Techniques Springer US. pp. 219- 263 ,(1996) , 10.1007/978-1-4613-1337-3_6
H.S. Sawhney, S. Ayer, M. Gorkani, Model-based 2D&3D dominant motion estimation for mosaicing and video representation international conference on computer vision. pp. 583- 590 ,(1995) , 10.1109/ICCV.1995.466886
Michal Irani, Stephen Hsu, P. Anandan, Mosaic-based video compression IS&T/SPIE's Symposium on Electronic Imaging: Science & Technology. ,vol. 2419, pp. 242- 253 ,(1995) , 10.1117/12.206363
Laura Teodosio, Walter Bender, Salient video stills: content and context preserved acm multimedia. pp. 39- 46 ,(1993) , 10.1145/166266.166270
R.G. Kermode, A.B. Lippman, Coding for content: enhanced resolution from coding international conference on image processing. ,vol. 3, pp. 3460- ,(1995) , 10.1109/ICIP.1995.537671
D. Hepper, Efficiency analysis and application of uncovered background prediction in a low bit rate image coder IEEE Transactions on Communications. ,vol. 38, pp. 1578- 1584 ,(1990) , 10.1109/26.61399