Method and device for scalable video transcoding

1. A method of modifying data in an input coded video signal for generating an output video signal, each video signal corresponding to a sequence of coded video frames, said method comprising at least:

an error decoding step for delivering a decoded data signal from a current input coded video frame,

a re-encoding step for delivering an output video frame, carried by said output video signal, from a modified signal, said modified signal resulting from an adding sub-step between a motion-compensated signal and said decoded data signal,

a reconstruction step for delivering a coding error of said output video frame,

a motion-compensation step for delivering said motion-compensated signal from a stored coding error of a previous output video frame, characterized in that said method comprises:

a first switching step inserted between said re-encoding step and said reconstruction step for activating said reconstruction step,

a second switching step inserted between said motion-compensation step and said adding sub-step for activating said motion-compensation step, said switching steps being controlled independently by a control strategy for defining a scalable method defining four architectures for modifying said input coded video signal;

wherein the control strategy of said first switching step, performed at the macroblock level, comprises:

a first prediction sub-step of the energy of said coding error, said coding error energy being predicted for all re-encoded macroblocks of an output video frame,

a first selection sub-step for identifying, among macroblocks defining said coding error, a first set of macroblocks having a predicted coding error energy below a given threshold, the coding error of said first set of macroblocks being set to a zero value so as to result in a modified stored coding error,

a second selection sub-step for identifying, among macroblocks defining said coding error, a second set of macroblocks having a predicted coding error energy above said given threshold,

said first switching step activating the reconstruction step for said second set of macroblocks.

2. A method of modifying data as claimed in claim 1 , wherein the control strategy of said second switching step, performed at the macroblock level, comprises:

a second prediction sub-step of the energy of said motion-compensated signal, said energy being predicted at the macroblock level,

a third selection sub-step for identifying, within said motion-compensated signal, a third set of macroblocks having a predicted energy above said given threshold,

said second switching step activating the motion-compensation step for each macroblock defining said decoded data signal, which motion-compensated macroblock belongs to said third set of macroblocks.

3. A method of modifying data as claimed in claim 1 , wherein the first prediction sub-step of the energy of said coding error consists in an energy calculation of a signal obtained after subtraction of said output video signal multiplied by a quantization factor from said modified signal.

4. A method of modifying data as claimed in claim 2 , wherein the second prediction sub-step of the energy of said motion-compensated signal consists in a weighted average of four previously predicted coding error energy values relating to four adjacent macroblocks which define said modified coding error.

5. A method of modifying data as claimed in claim 2 , wherein the motion-compensation step, when activated by said second switching step, is performed on the basis of said modified stored coding error.

6. A method of modifying data as claimed in claim 1 , wherein the first switching step is activated so as to obtain a decreasing number of reconstructed macroblocks over a GOP.

7. A method of modifying data as claimed in claim 1 , wherein the switching steps are controlled at the frame level, a given input coded video frame being modified in accordance with one of said four architectures.

8. A transcoding device for modifying data in an input coded video signal for generating an output video signal, each video signal corresponding to a sequence of coded video frames, said device comprising at least:

decoding means for delivering a decoded data signal from a current input coded video frame,

re-encoding means for delivering an output video frame, carried by said output video signal, from a modified signal, said modified signal resulting from addition means adding a motion-compensated signal to said decoded data signal,

reconstruction means for delivering a coding error of said output video frame,

motion-compensation means for delivering said motion-compensated signal on the basis of a stored coding error of a previous output video frame, characterized in that said transcoding device comprises:

first switching means inserted between said re-encoding means and said reconstruction means for activating said reconstruction means,

second switching means inserted between said motion-compensation means arid said addition means for activating said motion-compensation means,

said switching means being controlled independently by control means for defining a scalable transcoding device defining four architectures for modifying said input coded video frames,

wherein the control means of said first switching means comprise:

first prediction means for predicting the energy of said coding error, said coding error energy being predicted for all re-encoded macroblocks of an output video frame,

first selection means for identifying, among macroblocks defining said coding error, a first set of macroblocks having a predicted coding error energy below a given threshold, the coding error of said first set of macroblocks being set to a zero value so as to result in a modified stored coding error,

second selection means for identifying, among macroblocks defining said coding error, a second set of macroblocks having a predicted coding error energy above said given threshold,

said first switching means activating the reconstruction means for said second set of macroblocks.

9. A transcoding device for modifying data as claimed in claim 8 , wherein the control means of said second switching means comprise:

second prediction means for predicting the energy of said motion-compensated signal, said energy being predicted at the macroblock level,

third selection means for identifying, within said motion-compensated signal, a third set of macroblocks having a predicted energy above said given threshold,

said second switching means activating motion compensation means for each macroblock defining said decoded data signal, which motion-compensated macroblock belongs to said third set of macroblocks.

10. A transcoding device for modifying data as claimed in claim 8 , wherein said modified coding error is stored in a memory whose capacity is proportioned for storing only reconstructed macroblocks.

11. A set-top box product for modifying data in an input coded video coded according to the MPEG-2 video standard, said set-top box product implementing steps and sub-steps of the method as claimed in claim 1 .

12. A computer program product for a transcoding device for modifying data in a coded video signal, which product comprises a set of instructions which, when loaded into said device, causes said device to carry out any method as claimed in claim 1 .