HDTV encoder with forward estimation and constant rate motion vectors

• US 5,144,423 A
• Filed: 12/11/1990
• Issued: 09/01/1992
• Est. Priority Date: 12/11/1990
• Status: Expired due to Term

• Current Assignee: American Telephone & Telegraph Company (AT&T, Inc.)
• Original Assignee: AT&T, Inc.
• Inventors: Safranek, Robert J., Westerink, Peter H., Netravali, Arun N., Matthews, Kim N., Petajan, Eric D., Knauer, Scott C.
• Primary Examiner(s): Peng, John K.
• Application Number:
  US07/626,356
• Time in Patent Office: 630 Days
• Field of Search: 358/105, 358/133, 358/140, 358/141, 358/135
• US Class Current: 375/240.17
• CPC Class Codes: H04N 19/39 involving multiple description coding [MDC], i.e. with separate layers being structured as independently decodable descriptions of input picture data
• CPC Class Codes: H04N 19/51 Motion estimation or motion compensation
• CPC Class Codes: H04N 19/523 with sub-pixel accuracy
• CPC Class Codes: H04N 19/567 Motion estimation based on rate distortion criteria
• CPC Class Codes: H04N 19/61 in combination with predictive coding
• CPC Class Codes: H04N 19/94 Vector quantisation

Abstract

A motion compensated encoder where motion vectors are selected based on the prediction error generated in localized areas of the encoded image and based on an available bit budget. The motion vectors are created by dividing the image into blocks of two sizes and by considering the best mix of large and small size blocks, and their associated motion vectors, that minimize the overall prediction error, within the constraints of the bit budget. For convenience, the image division is arranged so that a given number of small sized blocks forms one large sized block (e.g. 16:1). Also, the block sizes are arranged so that employing only large sized blocks does not exceed the given bit budget, while employing only the small sized blocks does exceed the given bit budget.

Claims

1. An HDTV encoder comprising:

a forward estimation block responsive to applied image frame pixel signals, I(t), which includes

means, responsive to said image frame pixel signals, for developing a plurality of different types of control signals within a chosen processing time interval, T, and

means for delaying said applied image frame signals by said chosen processing time interval to develop a delayed signal, I(t-T); and

an encoding loop responsive to said control signals and to said delayed signal I(t-T) for developing quantized code signals for said applied image frame pixel signals.

2. The encoder of claim 1 wherein said means for developing control signals includes means for developing a frame standard deviation control signal which corresponds to a measure of variation in the signal I(t).

3. The encoder of claim 1 wherein said means for developing control signals includes means for developing a frame-mean control signal which corresponds to the average, signal level of I(t).

4. The encoder of claim 1 wherein said means for developing control signals includes means for developing leak factor control signal which relates to a variability of signal I(t) as compared to signal I(t-T).

5. The encoder of claim 1 wherein said applied image frame pixel signals are divisible into different sized blocks of pixel signals, and said means for developing control signals includes means for developing motion vector signals, which signals specify motion to be imparted to the different sized blocks of pixel signals in order to estimate signal I(t).

6. The encoder of claim 5, wherein said encoding loop includes motion compensation means responsive to said motion vectors developed by said means for developing control signals, that specify motion to be imparted to the different sized blocks of pixel signals to estimate signal I(t).

7. The encoder of claim 5 wherein said means for delaying said applied image frame signals by said chosen processing time interval includes means responsive to said I(t-T) and to said motion vectors for developing a displaced frame, I'"'"'(t) signal that estimates said signal I(t).

8. The encoder of claim 1 wherein said means for developing control signals includes means for developing motion vector signals with which to predict signal I(t), which signals specify motion to be imparted to blocks of I(t-T) that combine to form I(t-T), where the blocks are of two sizes:

one having a smaller size, and the other having a larger size.

9. The encoder of claim 8 where said means for developing motion vector signals selects a mix of smaller size blocks and larger size blocks that, together, requires a number of bits to specify which is not larger than a preselected number of bits.

10. The encoder of claim 8 where said means for developing motion vector signals selects smaller size blocks and larger size blocks in a mix of motion vectors that requires a number of bits to specify which does not exceed a preselected number of bits by more than a given value.

11. The encoder of claim 10 where said means for developing motion vectors selects said mix to minimize an error measure in the estimation of signal I(t) from signal I(t-T).

12. The encoder of claim 8 wherein said means for developing motion vectors includes:

first means, responsive to I(t) and to I(t-T), for developing coarse estimates of motion vectors, for specifying motion to be imparted to a given sized block of I(t-T) to estimate signal I(t); and

second means, responsive to I(t), I(t-T) and to said coarse estimates of said first means, for specifying motion to be imparted to said given sized blocks of I(t-T) and to collections of smaller sized blocks of I(t-T), where each collection of smaller sized blocks form said given sized block, for developing finer estimates of motion vectors for specifying motion to be imparted to the given sized blocks or to the smaller sized blocks to estimate said signal I(t).

13. The encoder of claim 12 further including third means for interpolating said finer estimates to sub-pixel accuracy.

14. The encoder of claim 11 further including means for variable length encoding of the selected mix of motion vectors.