Title :
ROI centered compression by Adaptive Quantization for sports video
Author :
Kim, Jong-Yun ; Yi, Chun-Ho ; Kim, Tae-Yong
Author_Institution :
GSAIM, Chung-Ang Univ., Seoul, South Korea
Abstract :
In this paper, we propose a region of interest (ROI) centered compression method for sports video. First we track a center of ROI using dynamic Kalman filter (DKF) algorithm that we proposed previously, determine ROI macro blocks (MBs) and finally quantize each MB adaptively with rate control. Experimental results show that our proposed algorithm presents increased clearness in the ROI than uniform quantization methods while maintaining stable bit-rate.
Keywords :
Kalman filters; data compression; sport; video coding; ROI macro blocks; adaptive quantization; dynamic Kalman filter algorithm; region of interest centered compression method; sports video; Digital multimedia broadcasting; Discrete cosine transforms; Games; Measurement errors; Object detection; Quantization; Robustness; Target tracking; Video compression; Watches;
Conference_Titel :
Consumer Electronics (ICCE), 2010 Digest of Technical Papers International Conference on
Conference_Location :
Las Vegas, NV
Print_ISBN :
978-1-4244-4314-7
Electronic_ISBN :
978-1-4244-4316-1
DOI :
10.1109/ICCE.2010.5418804
