Preprocessing method applied to textures of arbitrarily shaped objects

• US 6,768,495 B2
• Filed: 02/01/2002
• Issued: 07/27/2004
• Est. Priority Date: 02/06/2001
• Status: Expired due to Fees

• Current Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Original Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Inventors: Valente, Stephane Edouard
• Primary Examiner(s): Bella, Matthew C.
• Assistant Examiner(s): Cunningham, G. F.
• Application Number:
  US10/061,084
• Publication Number:
  US 20020168114A1
• Time in Patent Office: 907 Days
• Field of Search: 345/427, 345/582, 345/586, 345/592, 345/615, 345/619, 345/654, 345/FOR.110, 345/FOR.111, 345/FOR.119, 345/FOR.211, 345/FOR.214, 348/128, 348/394.1, 348/395.1, 348/FOR.133, 348/FOR.134, 348/FOR.138, 382/108, 382/169, 382/243, 382/248, 382/250, 382/276-283, 382/302
• US Class Current: 345/619
• CPC Class Codes: H04N 19/21 with binary alpha-plane coding for video objects, e.g. context-based arithmetic encoding [CAE]
• CPC Class Codes: H04N 19/649 the transform being applied to non rectangular image segments

Abstract

The invention relates to a method of preprocessing data that correspond to arbitrarily shaped objects and comprise for each object a texture part and an object mask, said method comprising, for each object plane associated to an object, the steps of:

(1) partitioning the object plane into bidimensional blocks;

(2) introducing in each block a set of basis vectors chosen in order to express an estimate of the original pixel values as a linear combination of said basis vectors;

(3) defining a cost function ψ to measure the distortion between the original expression of the pixel values and said estimate of this original expression;

(4) finding the coefficients allowing to minimize said cost function ψ, said finding step itself comprising the following operations: initialization, extraction of the basis vectors restricted to the opaque pixels and calculation of projection coefficients, iterations, and interruption of said iterations according to a predetermined criterion.

Claims

1. A method of preprocessing input data corresponding to picture elements (pixels) that represent arbitrarily shaped objects, said input data comprising for each object a texture part, corresponding to the values of the pixels of said object, and an object mask, subdividing said input data into a first and a second subset of data respectively corresponding to fully or partially opaque pixels and to transparent pixels in said texture part, said preprocessing method being provided for determining DCT (discrete cosine transform) coefficients corresponding to said opaque pixels and comprising for each considered object the steps of:

(1) partitioning the object plane into bidimensional blocks;

(2) introducing in the picture area defined by said block a set of basis vectors chosen in order to express an estimate of the original pixel values as a linear combination of said basis vectors;

(3) defining a cost function ψ

to measure the distortion between the original expression of the pixel values and said estimate of this original expression;

(4) finding the coefficients allowing to minimize the cost function ψ

;

characterized in that;

(a) said cost function Ψ

is given by a relation of the type;

$\mathrm{\psi <br><br>}=\left(f_{\mathrm{opaque}},\underset{i=1}{\overset{64}{\sum <br><br>}}<br><br>\text{\hspace{1em}<br><br>}<br><br>c_i<br><br>b_{\mathrm{opaque}<br><br>\left(i\right)}\right)$

 

where f is the column-vector of the pixels of the concerned block, ((b_i), iε

(1 to

     64)) are the basis vectors of a 8×

8 DCT, f_opaque is the restriction of f to the opaque pixels of said block, ((b_opaque), iε

(1 to

     64)) are the restriction of said basis vectors to the location of the opaque pixels of the block, and $\underset{i=1}{\overset{64}{\sum <br><br>}}<br><br>\text{\hspace{1em}<br><br>}<br><br>c_i<br><br>b_{\mathrm{opaque}<br><br>\left(i\right)}$

is called the reconstruction of f_opaque;

(b) said finding step itself comprises the following operations;

initialization of the following parameters, including;

iteration parameter k=0;

initial estimation of f_opaque^E=0;

initial reconstruction coefficients c_i⁰=0;

extraction of the basis vectors restricted to the opaque pixels and calculation of the projection coefficients;

2. A preprocessing method according to claim 1, characterized in that the cost function is defined by the relation:

$\mathrm{\psi <br><br>}={<br><br>f_{\mathrm{opaque}}-\underset{i=1}{\overset{64}{\sum <br><br>}}<br><br>\text{\hspace{1em}<br><br>}<br><br>c_i<br><br>b_{{\mathrm{opaque}}_{\left(i\right)}}<br><br>}^2$

and the sub-step is provided for finding the index i* of the basis vector which captures the maximum energy of the residual $f_{\mathrm{opaque}}-\underset{i=1}{\overset{64}{\sum <br><br>}}<br><br>\text{\hspace{1em}<br><br>}<br><br>c_i<br><br>b_{{\mathrm{opaque}}_{\left(i\right)}}<br><br>\text{;<br><br>}$ 

i*=arg.max∥

p_i^k·

b_opaque(i)∥

²for i=1 to 64.

3. A method of coding input data corresponding to the texture of arbitrarily shape objects, said coding method including at least a DCT transform of said texture input data, a quantization of the coefficients obtained from said transform, a differential prediction of the data to be coded, and a variable length encoding operation of said quantized coefficients, characterized in that said DCT coefficients are the coefficients c_i^k obtained by implementation of a preprocessing method according to claim 1 and corresponding to the opaque pixels of the arbitrarily shaped objects.

4. An encoding device for coding input data corresponding to the texture of arbitrarily shape objects, said encoding device including at least means for implementing a DCT transform of said texture input data, a quantization of the coefficients obtained from said transform, a differential prediction of the data to be coded, and a variable length encoding operation of said quantized coefficients, characterized in that said DCT coefficients are the coefficients c_i^k obtained by implementation of a preprocessing method according to claim 1 and corresponding to the opaque pixels of the arbitrarily shaped objects.