« ZurückWeiter »
(12) United States Patent
Reichel et al.
(io) Patent No.: (45) Date of Patent:
US 7,881,370 B2 Feb. 1, 2011
(21) (22) (86)
(87) (65) (30)
METHOD OF SELECTING AMONG N
SPATIAL VIDEO CODECS THE OPTIMUM
CODEC FOR A SAME INPUT SIGNAL
Inventors: Julien Reichel, Renens (CH); Francesco
Ziliani, Lausanne (CH)
Assignee: Visiowave S.A., Ecublens (CH)
Notice: Subject to any disclaimer, the term of this patent is extended or adjusted under 35 U.S.C. 154(b) by 1314 days.
§ 371 (c)(1), (2), (4) Date:
PCT Pub. No.:
10/539,207 Dec. 9, 2003 PCT/IB03/05852
Mar. 29, 2006 WO2004/056124 PCT Pub. Date: Jul. 1, 2004
Prior Publication Data US 2006/0215751 Al Sep. 28, 2006
Foreign Application Priority Data
Dec. 17,2002 (EP) 02406109
(51) Int. CI.
H04N 7/12 (2006.01)
(52) U.S. CI 375/240.03; 375/240.01;
375/240.18; 375/240.19; 375/240.2
(58) Field of Classification Search 375/240.01,
375/240.03,240.18-240.2 See application file for complete search history. (56) References Cited
U.S. PATENT DOCUMENTS
5,047,852 A * 9/1991 Hanyu et al 375/240.2
The invention relates to a method of selecting among N "Spatial Video CODECs" where N is an integer number greater than 1, the optimum "Spatial Video CODEC" for a same input signal I. "Spatial Video CODEC" is understood as the combination of any transform of the input signal, followed by a quantization of the transform coefficients and a corresponding entropic coder. Firstly from all the N "Spatial Video CODECs" for the same input signal I and a same quality parameter Q, the rate R and the distortion measures D are obtained. Q is an integer value between 0 and 100, defined by any rate-distortion algorithm to provide a compression of the input sequence with constant rate or with constant distortion. Further an optimality criterion is calculated. The optimality criterion is defined as the minimization of the value Ln=f(Rn, Dn) calculated for all the n from 1 to N. n is the index of the "Spatial Video CODEC", where f(Rn,Dn) is a function of Rn and Dn. The function f may be defined as as f(Rn,Dn)=Rn+ lDn, , 1 being the Lagrange multiplier that weights the relative influence of the rate R and of the distortion D.
7 Claims, 5 Drawing Sheets
5,394,473 A * 2/1995 Davidson 704/200.1
5,778,192 A 7/1998 Schuster et al.
5,959,560 A * 9/1999 Saidetal 341/107
Mallat S. et al., "Analysis of Low Bit Rate Image Transform Coding", IEEE Lransactions on Signal Processing, IEEE, Inc., New York, US, vol. 46, No. 4, Apr. 1, 1998, pp. 1027-1042.
* cited by examiner