GB2286500A - Wide to standard format motion compensated video signal processor - Google Patents

Abstract

The motion vector processor determines motion vectors representative of motion within successive images and is also responsive to a default motion vector representative of motion of the images themselves (pan and scan). Alternatively, or in addition, block matching for determining motion vectors representative of movement within the first images can be responsive to an image offset representing relative movement between a pair of first images to offset one image of the pair by half the offset in a first direction and to offset the other image of the pair by half the offset in the opposite direction. The apparatus enables motion compensation to be performed on moving a letterbox representation of a high resolution image within a normal resolution video image (see fig. 6) and on panning and scanning an edge cropped representation of a high resolution image within a normal resolution video image. <IMAGE>

Description

MOTION COMPENSATED VIDEO SIGNAL PROCESSING This invention relates to motion compensated video signal processing.

Motion compensated video signal processing is used in applications such as television standards conversion, film standards conversion and conversion between video and film standards.

Motion compensated video signal processing requires powerful and complex processing apparatus to carry out the very large number of calculations required to generate and process motion vectors for each pair of input images. This is particularly true if the images are in a high definition format, or if the processing is to be performed on an input video signal to produce an output video signal in real time, in which case multiple sets of identical apparatus may be operated in parallel in order to generate sets of motion vectors for each output image in the time available (e.g. an output field period).

In a motion compensated television standards converter, such as the converter described in the British published patent application number GB-A-2 231 749, pairs of successive input images are processed to generate sets of motion vectors representing motion within the pair of input images. The processing is carried out on discrete blocks of the images, so that each motion vector represents the inter-image motion of the contents of a respective block.

Each set of motion vectors is then supplied to a motion vector reducer which derives a subset of the set of motion vectors for each block. The subset is then passed to a motion vector selector which assigns one of the subset of motion vectors to each picture element (pixel) in each block of the image. The selected motion vector for each pixel is supplied to a motion compensated interpolator; the interpolator operates on progressive-scan converted versions of the input images to interpolate successive output images, taking into account the motion between the input images.

Motion compensated video signal processing such as that described in GB-A-2 231 749 enables inter-image motion compensation. However, it assumes that the mapping between the input images and the output images is constant.

In accordance with one aspect of the invention, there is provided a video signal processing apparatus for processing a sequence of first images to generate a sequence of second images comprising a motion compensator for compensating for motion within successive first images, characterised by means for compensating for motion resulting from changes in a mapping of respective first images with respect to said second images.

Thus, a video signal processing apparatus in accordance with the invention enables compensation for motion resulting from changes in a mapping of successive first images onto the second images as well as compensation for motion within the first images. It avoids the generation of artifacts which might otherwise form at the edges of the first images as viewed in the resulting second images when the mapping between the first and second images is changed, for example when a letterbox representation of a high resolution image is moved within a normal resolution image, or a cropped edge representation of a high resolution image is panned or scanned, or when a scaled version of an input image is translated within an output image.

In one preferred embodiment the apparatus comprises means for defining changes to a mapping of successive first images with respect to said second images including generating a default motion vector representative of relative movement between respective first images with respect to said second images as a result of said mapping changes.

Preferably, in this embodiment said motion compensator comprises an interpolator for interpolating between successive pairs of first images and a motion vector processor for determining motion compensation vectors for controlling said interpolator. Said motion vector processor determines inter-image motion vectors representative of motion within successive first images and is also responsive to said default motion vector for determining said motion compensation vectors.

Preferably also, said motion vector processor comprises a motion vector estimator for estimating motion vectors for image areas (e.g., blocks) in pairs of successive first images, a motion vector reducer for reducing the number of motion vectors for each image area and a motion vector selector for selecting motion vectors to form said motion compensation vectors, said motion vector reducer being responsive to said inter-image motion vectors and also to said default vector as an additional vector for at least certain image areas.

In order, specifically to avoid artifacts at the edges of a representation of first images within respective second images where said first images are mapped onto an active region of said second images, the non-active remainder of said second images being blank, said motion vector reducer is preferably responsive to said default vector only in image areas corresponding to said non-active region and portions of said active region bordering on said non-active region of said second images.

Alternatively or in addition said apparatus preferably comprises means for defining changes to a mapping of said first images with respect to said second images including means defining an image offset vector for a pair of successive input images representative of relative movement of said pair of first images with respect to said second images as a result of said mapping changes.

Preferably said apparatus comprises a block matching processor for matching blocks of a pair of successive first images for determining motion vectors representative of movement within said first images, wherein said block matching processor is responsive to said offset vector relatively to offset said pair of images for said block matching to compensate for said relative movement of said pair of first images, and means for adding said offset vector to a set of vectors resulting from vector estimation and/or reduction processing subsequent to said block matching.

Said block matching processor can be arranged simply to offset one image of said pair by half of said offset vector in a first direction and to offset the other image of said pair by half of said offset vector in an opposite direction for said block matching.

Alternatively, especially where the relative movement of the images may be accelerating of decelerating, said block matching processor can be arranged to offset one image of said pair by a fraction X/A of said offset vector in a first direction, where X is in the range 0 to A, and to offset the other image of said pair by the remainder (A-X)/A of said offset vector in an opposite direction for said block matching. It is not necessary for an exact ratio to be employed. A coarse adjustment could alternatively be employed (e.g., a number of discrete alternative offsets).

The apparatus in accordance with the invention finds particular, but not exclusive application to converting high resolution video images to normal resolution video images, for example for vertical motion of a letterbox representation of a high resolution video image within a normal resolution video image with compensation for the vertical motion of the letterbox representation and/or for panning and scanning of an edge cropped representation of a high resolution video image within a normal resolution video image with compensation for horizontal motion of said cropped representation resulting from said panning and scanning. The invention also finds application to the translation of a scaled representation of first video images within second video images with compensation for motion resulting from said translation.

Particularly for video standards conversion, the apparatus preferably comprises means for receiving input video images, a downconverter for generating said first video images from said input video images wherein said first video images have a lower resolution than said input video images.

In particular examples of the invention the input video images may be high resolution and/or interlaced digital video images, for example with the second video images being normal resolution digital video images.

The invention also provides a video standards converter apparatus comprising a video signal processing apparatus as defined above. The invention also provides a television standards conversion apparatus, comprising apparatus as defined above for converting from 16:9 aspect ratio high resolution images to 4:3 aspect ratio normal resolution images.

The invention also provides a film standards conversion apparatus and/or film to television standards conversion apparatus comprising apparatus as defined above.

In accordance with a second aspect of the invention there is provided a method of processing a sequence of first images to generate a sequence of second images comprising steps of: compensating for motion within successive first images; and compensating for motion resulting from changes in a mapping of respective first images with respect to said second images.

Embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings, throughout which like parts are referred to by like references, and in which: Figure 1 is a schematic overview in block diagrammatic form of a motion compensated television standards conversion apparatus; Figure 2 is a schematic block diagram of a first embodiment of a motion compensated frame rate converter in accordance with the invention; Figure 3 is a schematic block diagram of a second embodiment of a motion compensated frame rate converter in accordance with the invention; Figure 4 is a schematic block diagram of a block matching processor of the second embodiment of Figure 3; Figure 5 is a schematic block diagram of a third embodiment of a motion compensated frame rate converter in accordance with the invention; Figures 6A, 6B and 6C schematically illustrate conversion operations which can be performed by a television standards conversion apparatus such as that illustrated in Figure 1; and Figures 7A, 7B and 7C schematically illustrate examples of image motion which can be compensated in apparatus in accordance with the invention.

Figure 1 is a schematic overview in block diagrammatic form of a motion compensated television standards conversion apparatus in accordance with the invention. The apparatus receives an input interlaced digital video signal 24 (e.g. an 1125/60 2:1 high definition video signal (HDVS)) and generates an output interlaced digital video signal 28 (e.g a 625/50 2:1 signal).

The input video signal 24 is first supplied to a spatial downconverter 20 which performs progressive scan conversion whereby the input video signals are converted to a progressive format (i.e. the 2:1 interlace is removed) in accordance with a motion adaptive interpolation scheme in which inter-frame interpolation is employed in static picture areas and inter-field interpolation is employed on moving material within the picture. This 1125 line, 60 Hz, 1:1 progressive format is then converted to 625 line. 60 Hz, 1:1 format which is output via a path 26 to a motion compensated frame-rate converter 22 where frame-rate conversion to 625 line, 50 Hz, 2:1 format with motion compensation is performed, the 625 line, 50 Hz, 2:1 format video signals being output at 28.

A control system 12, e.g. a suitably programmed personal computer, provides timing and other control information to the spatial down-converter 20 and the motion compensated frame-rate converter 22 via paths 16 and 18, respectively, in response to operator commands or command lists supplied at 14.

Figure 2 is a schematic block diagram of a first embodiment of a motion compensated frame-rate converter in accordance with the invention.

The data from the spatial down-converter 20 are input at 26 to a time base changer 30. The time base changer 30 determines the temporal position of each progressive scan frame of the output video signal, and selects the two progressive scan frames of the input video signal which are temporally closest to that output progressive scan frame for use in interpolating that output progressive scan frame. For each progressive scan frame. of the output video signal, the two input progressive scan frames selected by the time base changer are passed via path 50 to a system delay 32 where they are delayed by an appropriate time before being supplied via a path 54 to an interpolator 34 in which that output progressive scan frame is interpolated. A control signal indicating the temporal position of each output progressive scan frame with respect to the two selected input progressive scan frames, is supplied from the time base changer 30 via path 48 to the system delay 32 and path 52 to the interpolator 34.

The time base changer 30 can operate according to synchronisation signals associated with the input video signal, the output video signal, or both. In the case in which only one synchronisation signal is supplied, the timing of the progressive scan frames of the other of the two video signals is generated deterministically within the time base changer 30.

The pairs of progressive scan frames selected by the time base changer 30 to be passed via the path 50 to the system delay 32 are also passed via path 56 to a motion vector processor 36 for the generation of motion compensation vectors.

The motion vector processor 36 comprises a direct block matcher 38, a motion vector estimator 40, a motion vector reducer 42, a motion vector selector 44 and a motion vector post-processor 46. The pairs of input progressive scan frames are supplied first to the direct block matcher 38 which calculates correlation surfaces representing the spatial correlation between search blocks in the temporally earlier of the two selected input progressive scan frames and (larger) search areas in the temporally later of the two input progressive scan frames, The correlation surfaces output by the block matcher 38 are passed to the motion vector estimator 40. The motion vector estimator 40 detects points of greatest correlation in the correlation surfaces.

(The correlation surfaces actually represent the difference between blocks of the two input progressive scan frames; this means that the points of maximum correlation are in fact minima on the correlation surfaces, which are referred to as "minima"). From the detected minimum on each correlation surface, the motion vector estimator 40 generates a motion vector which is supplied to the motion vector reducer 42.

The motion vector estimator 40 can also be arranged to perform a confidence test on each generated motion vector to establish whether that motion vector is significant above the average data level, and associates a confidence flag with each motion vector indicative of the result of the confidence test. An example of a confidence test, known as the "threshold" test, is described (along with certain other features of the apparatus of Figure 1) in GB-A-2 231 749.

The motion vector reducer 42 operates to reduce the choice of possible motion vectors for each pixel of the output progressive scan frame, before the motion vectors are supplied to the motion vector selector 44. The output progressive scan frame is notionally divided into blocks of pixels, each block having a corresponding position in the output progressive scan frame to that of a search block in the earlier of the selected input progressive scan frames. The motion vector reducer 42 compiles a group of four motion vectors to be associated with each block of the output progressive scan frame, with each pixel in that block eventually being interpolated using a selected one of that group of four motion vectors.

As part of its function, the motion vector reducer 42 counts the frequencies of occurrence of "good" motion vectors (i.e. motion vectors which pass the confidence test and the alias test, or which were re qualified as non-aliased), with no account taken of the position of the blocks of the input progressive scan frames used to obtain those motion vectors. The good motion vectors are then ranked in order of decreasing frequency. The most common of the good motion vectors which are significantly different to one another are then classed as motion vectors. Three motion vectors which pass the confidence test are then selected for each block of output pixels and are supplied, with a zero motion vector, to the motion vector selector 44 for further processing.

The three selected motion vectors are selected in a predetermined order of preference from: (i) the motion vector generated from the corresponding search block (the "local" motion vector"); (ii) those generated from surrounding search blocks ("neighbouring" motion vectors): and (iii) the global motion vectors.

The zero motion vector, or stationary motion vector is normally supplied to ensure that the motion vector selector 44 is not forced into applying a motion vector representing motion to a stationary pixel.

The motion vector selector 44 also receives as inputs the two input progressive scan frames which were selected by the time base changer 32 and used to calculate the motion vectors. These progressive scan frames are suitably delayed in the system delay 32 so that they are supplied to the motion vector selector 44 via path 60 (along with the control signal indicating the temporal position of each output progressive scan frame with respect to the two selected input progressive scan frames on path 62) at the same time as the vectors derived from them. The motion vector selector 44 supplies an output comprising one motion vector per pixel of the output progressive scan frame. This motion vector is selected from the four motion vectors for that block supplied by the motion vector reducer 42.

The vector selection process involves detecting the degree of correlation between test blocks of the two input progressive scan frames pointed to by a motion vector under test. The motion vector having the greatest degree of correlation between the test blocks is selected for use in interpolation of the output pixel.

The motion vectors are supplied via a path 46 from the vector selector 44 to interpolator 34. Using the motion vectors, the interpolator 34 interpolates an output field from the corresponding two progressive scan frames selected by the time base changer 30, taking into account any image motion indicated by the motion vectors currently supplied to the interpolator 34.

Pixels from the two selected progressive scan frames supplied to the interpolator 34 are combined in relative proportions depending on the temporal position of the output progressive scan frame with respect to the two input progressive scan frames (as indicated by the control signal t), so that a larger proportion of the nearer input progressive scan frame is used. The output 28 of the interpolator 34 is in the form of interlaced 625 line, 50 Hz, 2:1 format video signals.

Further explanation of motion vector estimation, reduction, and selection processes suitable for use in the present invention are described in published UK patent application GB-A-2248361.

The apparatus described above and in, for example, GB-A-2248361, provides inter-image motion compensation based on movement within the input images. It assumes that there is a constant mapping between the input images and the output images.

Figure 6 illustrates various mappings of a high definition television image with a 16:9 aspect ratio onto a normal definition television image with a 4:3 aspect ratio.

To the left of the arrows in Figures 6A, 6B and 6C high definition television images 70 with a 16:9 aspect ratio are illustrated.

To the right of the arrow in Figure 6A a letterbox representation 72 (in dotted lines) of the high definition image in a normal 4:3 definition television image 74 (in solid lines) is shown. Normally, the area above and below the letterbox representation is blank, although it can be used for the display, for example, of sub-titles.

To the right of the arrow in Figure 6B an edge-cropped representation 76 (in dotted lines) of the high definition image in a normal 4:3 definition television image 78 (in solid lines) is shown.

To the right of the arrow in Figure 6C a scaled representation 80 (in dotted lines) of the high definition image in a normal 4:3 definition television image 82 (in solid lines) is shown.

Typically, one of the above representations of the high definition image may be used. It will be appreciated that other representations such as a squeezed or stretched representation may also be used. However, in the normal course of events, the representation used is constant.

In some situations, however, it may be desirable to move the representation of the high definition image with respect to the lower definition image Figures 7A, 7B and 7C represent different translational movements.

Comparing the left and right hand sides of Figure 7A illustrates a situation where the letter box representation 72 of the high definition image is moved upwardly to 73 within the normal definition image 74.

Comparing the left and right hand sides of Figure 7B illustrates a situation where the edge cropped representation 76 of the high definition image is moved to the right to 77 with respect to the normal definition image 78. This type of function is referred to as pan and scan.

Comparing the left and right hand sides of Figure 7C illustrates a situation where the scaled representation 80 of the high definition image is moved diagonally downwards and to the left to 81 in the normal definition image 82.

It will be appreciated that these are merely examples of possible image movements. These movements could be done suddenly, but it would be advantageous if it could be done smoothly as a continual motion. In the latter case, the apparatus as described above would cause the generation of artifacts due to edge effects.

In the example of moving the letterbox image 72, 73 (Figure 7A), the artifacts (e.g., judder) would arise from interpolation by the interpolator on the basis of pixel information at or near the upper 'U' and lower 'L' edges of the active region corresponding to the representation of the high definition image 72, 73 and blank pixel information from the non-active or blank regions at the top and bottom of the active region. In the example of moving the edge cropped image (Figure 7B), the artifacts would arise from different information being available in the pairs of images at the left and right hand edges which would introduce noise. In the example of moving the scaled image (Figure 7C), the artifacts would arise from the different information available in the pairs of images at all the image edges also producing noise.

Control of the movement of the representations of the high definition images within the normal definition images can as such be provided in a conventional manner as is known for example, in digital video effect machines, by means of user commands or a command list input at 14 to the control system 12. In the preferred embodiment the control system is a personal computer programmed with a user interface of the type known in digital video effects machines and allowing the input, by means of conventional user input devices, of commands for controlling image manipulations.

Thus the personal computer 12 is programmed to respond to user instructions to determine a smooth movement of the high definition image between a first mapping onto the normal definition image and a second mapping onto the normal definition image (e.g. from 72 to 73 in Figure 7A, or 76 to 77 in Figure 7B or 80 to 81 in Figure 7C). One or more options for specifying the desired changes to the mappings of the input images onto the output images can be provided. For example, the intended final position of the input image in the output image can be specified and the time for the movement from the input to the output position specified. Alternatively, a standard speed of movement can be used, or a speed of movement specified. As a further alternative.

real-time control of the movement could be provided by means of a tracking device (e.g. a mouse).

The control system is arranged to compute, from the input commands, an image motion vector representative of the amount and direction of the movement between successive pairs of input images.

The image motion vectors are represented as V(a), V(b) and V(c) in Figures 7A, 7B and 7C, respectively.

In the first specific embodiment of the invention illustrated in Figure 2, these image motion vectors are supplied as a default vector to the vector reducer 42, which forms part of the motion vector processor 36, via path 66 which forms part of the control paths 18 from the control system to the motion compensated frame-rate converter 22.

This default vector can then be used selectively by the vector reducer 42 as a substitute for the zero motion vector mentioned above which is used for the static images.

The default vector on path 66 can be used as a substitute for the zero motion vector for all image blocks. However, as an alternative, the vector reducer could additionally be provided with control signals on the path 66 from the control system 12 whereby the zero motion vector is substituted for a zero motion vector only for those blocks outside an active area (i.e. outside the area of the normal definition image occupied by the representation of the high definition image) and those blocks within the active area but close to the boundary with the non-active area.

An effect of the first embodiment can be to reduce the available vector range available for the motion compensation vectors for interimage motion as a result of replacing the zero vector with the non-zero default vector. A typical range available is +/-24 pixels horizontally and +/-16 pixels vertically, determined by the number of bits which can be processed by the block matcher 38 and motion vector estimator 40 hardware. If a number of the available bits is reserved by the nonzero default vector (determined by the maximum speed of movement of the images) then the maximum speed of the inter-image motion which can be accommodated is reduced.

Figure 3 relates to a second specific embodiment of the invention, which does not reduce the available vector range for interimage motion. In this embodiment the motion of successive images is removed prior to the vector estimation process. This ensures that the full range of inter-image motion tracking can be employed in the vector estimator for inter-image motion (i.e. the content of the images) and none of this range need be used up following the intra-image motion (i.e. the motion of the images themselves).

In this embodiment the frame pair used for motion vector estimation in the block matcher 38' is offset in dependence upon the amount of the imposed image motion to realign them. Thus, for example, if the imposed motion is four pixels upwards (see Figure 7A) at a constant velocity then the previous frame would be offset by two pixels upwards and the next frame by two pixels downwards.

This would effectively remove the imposed motion for each output progressive scan frame. In this embodiment the imposed "global" motion vector is effectively used to pre-motion compensate successive images.

Thus, in this embodiment the image motion represented by the offset vector 68 (e.g. pan and scan motion) is removed from the progressive scan frames fed to the vector estimator 40. However, it still remains imposed on the progressive scan frames fed to the vector selector 44 and interpolator 34. Accordingly, the offset vector must be added to the vectors produced by the vector estimator 40 and vector reducer 42.

Specifically, the vector representative of the relative motion between the input and output images as computed in the system controller is used to compute offset vectors which are supplied to the block matcher 38' on a path 68, which forms part of the control paths 18 from the control system to the motion compensated frame rate converter 22.

Figure 4 represents, in block diagrammatic form, the block matcher 38' of the embodiment of Figure 3. The block matcher 38' comprises a conventional block matcher 38 (as in the embodiment of Figure 2) with, in addition, two programmable delays 37 and 39 for respective ones of the pairs of selected progressive scan frames on paths 56. The delays applied by the programmable delays 37 and 39 are controlled a block match controller 35 in response to the offset vector supplied on path 68. The progressive scan frames of a pair delayed by a respective amounts to realign those frames before being passed to the block matcher 38 for processing in the normal way.

The block match controller 35 can be arranged, for example, simply relatively to advance the earlier progressive scan frame by half of the offset vector value and to retard the later progressive scan frame by half the offset vector value.

Alternatively, for example to adapt the motion compensation where the relative movement between the input and output images is accelerating or decelerating, an uneven adjustment of the positions of the pair of progressive scan images can be employed. For example, where the magnitude of the offset vector is A, the first image of s an exact ratio to be employed. A coarse adjustment could alternatively be employed (e.g., a number of discrete alternative offsets).

Accordingly, for constant velocity, X will typically equal A/2, although where the movement is not constant X can be adapted accordingly.

As mentioned above, it is necessary to add the offset vector to the vectors produced by the vector estimator 40 and vector reducer 44.

Advantageously this can be achieved by means of a vector adder 43 connected to the output of the vector reducer 42, where the values "a", "b" and "c" shown in Figure 3 are as follows in the preferred embodiment: a = Vxy where Vxy is a set of four vectors per block; c = V0 where V0 is the offset vector on line 68; and b = Vxy + VO.

In a third specific embodiment of the invention illustrated in Figure 5, the embodiments of Figures 2 and 3 are effectively combined so that motion of the image can be compensated by replacing the zero vector with the default vector, by offsetting the input images for block matching, or by a combination of the two techniques.

Thus there have been described video signal processing apparatus and methods for processing a sequence of first images to generate a sequence of second images including motion compensation for motion within successive first images for generating the second images and for motion resulting from changes in a mapping of successive first images onto the second images. Motion vectors representative of motion within successive images with a default motion vector representative of motion of the images themselves can be used to this effect. Alternatively, or in addition, block matching for determining motion vectors representative of movement within the first images can be responsive to an image offset representing relative movement between a pair of first images to offset one image of a pair in a first direction and to offset the other image of the pair in the opposite direction.

The invention has been particularly described in the context of television standards conversion and illustrates examples of motion compensation on moving a letterbox representation of a high resolution image within a normal resolution video image, on panning and scanning an edge cropped representation of a high resolution image within a normal resolution video image, and on translating a scaled high resolution image within a normal resolution image.

Although particular embodiments of the invention have been described, it will be appreciated that modifications and/or additions thereto are possible within the scope of the invention.

For example, although the invention has been specifically described in the context of a television standards converter, it will be appreciated that it also finds application to film standards conversion and/or film to television standards conversion, or indeed more generally to video image processing.

Also, although the invention has been described in the context of standards conversion from high to lower definition images, it is not limited thereto and finds application in other situations where a variable mapping between images is employed.

Claims (29)

1. Video signal processing apparatus for processing a sequence of first images to generate a sequence of second images comprising a motion compensator for compensating for motion within successive first images, characterised by means for compensating for motion resulting from changes in a mapping of respective first images with respect to said second images.

2. Apparatus according to claim 1, comprising means for defining changes to a mapping of successive first images with respect to said second images including generating a default motion vector representative of relative movement between respective first images with respect to said second images as a result of said mapping changes.

3. Apparatus according to claim 2, wherein: said motion compensator comprises an interpolator for interpolating between successive pairs of first images and a motion vector processor for determining motion compensation vectors for controlling said interpolator; and said motion vector processor determines inter-image motion vectors representative of motion within successive first images and is also responsive to said default motion vector for determining said motion compensation vectors.

4. Apparatus according to claim 3, wherein said motion vector processor comprises: a motion vector estimator for estimating motion vectors for image areas in pairs of successive first images; a motion vector reducer for reducing the number of motion vectors for each image area; and a motion vector selector for selecting motion vectors to form said motion compensation vectors, said motion vector reducer being responsive to said inter-image motion vectors and also to said default vector as an additional vector for at least certain image areas.

5. Apparatus according to claim 4, wherein: said first images are mapped onto an active region of said second images, the non-active remainder of said second images being blank; and said motion vector reducer is responsive to said default vector only in image areas corresponding to said non-active region and portions of said active region bordering on said non-active region of said second images,

6. Apparatus according to any one said preceding claims, comprising means for defining changes to a mapping of said first images with respect to said second images including means defining an image offset vector for a pair of successive input images representative of relative movement of said pair of first images with respect to said second images as a result of said mapping changes.

7. Apparatus according to claim 6, comprising a block matching processor for matching blocks of a pair of successive first images for determining motion vectors representative of movement within said first images, wherein said block matching processor is responsive to said offset vector relatively to offset said pair of images for said block matching to compensate for said relative movement of said pair of first images, and means for adding said offset vector to a set of vectors resulting from vector estimation and/or reduction processing subsequent to said block matching.

8. Apparatus according to claim 7, wherein said block matching processor is arranged to offset one image of said pair by half of said offset vector in a first direction and to offset the other image of said pair by half of said offset vector in an opposite direction for said block matching.

9. Apparatus according to claim 7, wherein said block matching processor is arranged to offset one image of said pair by a fraction X/A of said offset vector in a first direction, where X is in the range O to A, and to offset the other image of said pair by the remainder (A-X)/A of said offset vector in an opposite direction for said block matching.

10. Apparatus according to any one of the preceding claims for converting high resolution video images to normal resolution video images.

11. Apparatus according to claim 10, providing for vertical motion of a letterbox representation of a high resolution video image within a normal resolution video image with compensation for the vertical motion of the letterbox representation.

12. Apparatus according to claim 10, providing for panning and scanning of an edge cropped representation of a high resolution video image within a normal resolution video image with compensation for horizontal motion of said cropped representation resulting from said panning and scanning.

13. Apparatus according to claim 10, providing for translation of a scaled representation of a high resolution video image within a normal resolution video image with compensation for motion resulting from said translation.

14. Apparatus according to any one of the preceding claims, comprising means for receiving input video images and a down-converter for generating said first video images from said input video images, wherein said first video images have a lower resolution than said input video images.

15. Apparatus according to claim 14, wherein said input video images are digital high resolution and/or interlaced video images.

16. Video standards converter apparatus comprising a video signal processing apparatus according to any one of the preceding claims.

17. Television standards conversion apparatus, comprising apparatus according to any one of the preceding claims for converting from 16:9 aspect ratio high resolution images to 4:3 aspect ratio normal resolution images.

18. Film standards conversion apparatus, comprising apparatus according to any one of the preceding claims.

19. Apparatus for converting between film and television standards, comprising apparatus according to any one of the preceding claims.

20. A method of processing a sequence of first images to generate a sequence of second images comprising steps of: compensating for motion within successive first images; and compensating for motion resulting from changes in a mapping of respective first images with respect to said second images.

21. A method according to claim 20, comprising steps of: defining changes to a mapping of successive first images with respect to said second images; and generating a default motion vector representative of relative movement between respective first images with respect to said second images as a result of said mapping changes.

22. A method according to claim 21, wherein said motion compensation comprises steps of: determining inter-image motion vectors representative of image motion within successive first images; determining motion compensation vectors from said inter-image motion vectors and said default motion vector; and interpolating between successive pairs of first images in dependence upon said motion compensation vectors.

23. A method according to any one of claims 20 to 22, comprising steps of: defining changes to a mapping of said first images with respect to said second images; and defining an image offset vector for a pair of successive input images representative of relative movement of said pair of first images with respect to respective second images for said pair as a result of said mapping changes.

24. A method according to claim 23, comprising steps of: responding to said offset vector relatively to offset said pair of images for said block matching to compensate for said relative movement of said pair of first images; and adding said offset vector to a set of vectors resulting from vector estimation and/or reduction processing subsequent to said block matching.

25. A method according to claim 24, comprising the step of: offsetting one image of said pair by half of said offset vector in a first direction and offsetting the other image of said pair by half of said offset vector in an opposite direction for said block matching.

26. Apparatus according to claim 24, comprising the step of: offsetting one image of said pair by a fraction X/A of said offset vector in a first direction, where X is in the range 0 to A, and offsetting the other image of said pair by the remainder (A-X)/A of said offset vector in an opposite direction for said block matching.

27. Video signal processing apparatus substantially as hereinbefore described with reference to the accompanying drawings.

28. Video standards converter substantially as hereinbefore described with reference to the accompanying drawings.

29. A method of motion compensated video signal processing, the method being substantially as hereinbefore described with reference to the accompanying drawings.