JPH08205179A - Apparatus and method for video signal coding - Google Patents

Abstract

PURPOSE: To improve the picture quality of videos by specifying a predicted present frame by assigning each picture element value of the preceding frame corresponding to one of the picture elements of the present frame as one picture element value of the present frame. CONSTITUTION: Selected characteristic points from a characteristic point selecting block 210 are inputted to a characteristic point motion vector detecting block 212 and first and second motion vector detecting blocks 214 and 22. At the block 212, first sets of motion vectors to the selected characteristic points are detected. When one characteristic point is received from the block 210, a characteristic block having a characteristic point at its center is retrieved from a frame memory through a line L12. After the motion vectors to all characteristic points are detected, the first sets of motion vectors are inputted to the block 214 and an entropy encoder 108 through a line L20. At the block 214, second sets of motions vectors contained in the present frame are specified by using the information from the block 210.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus and method for encoding a video signal, and more particularly to an apparatus and method for encoding a video signal using a pixel-by-pixel motion estimation technique.

[0002]

BACKGROUND OF THE INVENTION As is known, the transmission of discretized image signals can retain better image quality than analog signals. When an image signal consisting of a series of image "frames" is represented in digital form, especially in the case of high definition television, a considerable amount of data must be transmitted. However, since the usable frequency band of a normal transmission channel is limited, in order to transmit a large amount of digital data through the limited channel band,
It is necessary to compress and reduce the amount of transmitted data. Among various video compression techniques, the so-called hybrid coding technique is known to be the most efficient. This technique is a combination of probabilistic coding techniques and temporal and spatial compression techniques.

Most of this hybrid coding technique
Motion-compensated DPCM (differential pulse code modulation), two-dimensional DC
T (Discrete Cosine Transform), DCT coefficient quantization and V
LC (variable length coding) or the like is used. This motion compensation D
PCM is a method of estimating the motion of an object between the current frame and a previous frame, predicting the current frame according to the motion flow of the object, and generating a difference signal representing the difference between the current frame and its predicted value. . This method is described in, for example, Staffan Ericsson.
n) "Fixed and Adaptive Pr
edictorsfor Hybrid Predic
“Tive / Transform Coding”, IE
EE Transactions on Commun
ications , COM-33 , No. 12 (198
December 5), or "A Motion Compe" by Ninomiya and Otsuka.
Nested Interframe Coding
Scheme for Television Pic
tures ”, IEEE Transactions
on Communications , COM-30 ,
No. 1 (January 1982).

The two-dimensional DCT reduces spatial redundancy between video data and converts digital video data, for example, 8 × 8 pixel blocks into a set of transform coefficient data. This technique is described in Chen and Pratt, "Scene Adaptiv."
e Coder ", IEEE Transaction
Son Communication, COM-3
2, No-3 (March 1984). Quantizer, zigzag scan and VL
By processing in C, the amount of data to be transmitted can be effectively compressed.

More specifically, in this motion compensated DPCM, the current frame data is predicted from its corresponding previous frame data based on the motion estimation between the current frame and the previous frame. The motion thus estimated is explained by the two-dimensional motion vector representing the displacement of the pixel between the current frame and the previous frame.

There are various methods for estimating the pixel displacement of an object, but generally they can be classified into two types. One of them is a block unit method, and the other is a pixel unit motion estimation.

In the block-based motion estimation method,
The block in the current frame is compared with the block in the previous frame or the like to identify the best matching block. For the current frame to be transmitted, the interframe displacement vector for all blocks (representing the movement of blocks between frames) is estimated. However, in the block-based motion estimation method, a blocking effect is generated by partitioning blocks in the motion compensation process.
Occurs and all pixels in the block do not move in one direction, the estimates are incorrect, resulting in coarser image quality overall.

On the other hand, when the pixel-by-pixel motion estimation method is used, the displacement is obtained for each pixel. This method allows for more accurate estimation of pixel values, as well as scale changes (eg, zooming, which is a movement perpendicular to the image plane).
oming)) can be handled easily. However, in the pixel-by-pixel method, motion vectors have to be specified for each and every pixel, and in practice it is not possible to transmit all motion vectors to the receiver.

[0009]

SUMMARY OF THE INVENTION It is therefore a primary object of the present invention to provide an improved pixel-by-pixel motion estimation and compensation technique using the feature points and assisted quasi-feature points according to the present invention. is there.

[0010]

In order to achieve the above object, according to the present invention, a motion compensated video signal encoder is used, which is based on a current frame and a previous frame of a digital video signal. A video signal encoding device for specifying a predicted current frame, comprising: first pixel selecting means for selecting a plurality of pixels from pixels included in the previous frame; and between the current frame and the previous frame. A first motion vector selecting means for finding a first set of motion vectors, each of which includes a motion vector representing a motion to each of the selected pixels, and a motion vector included in the current frame using the first motion vector selection means. A second motion vector generating means for generating a second set of motion vectors for all the pixels, and one of the second set of motion vectors, in the current frame. First identifying means for identifying a preliminary predicted current frame by assigning each pixel value in the previous frame corresponding to one of the primes as the one pixel value in the current frame; A collective error region selecting means for searching for a collective error region by specifying a difference between the current frame and the preliminary predicted current frame, and selecting one pixel from pixels included in each of the collective error regions. Using two-pixel means, third motion vector selecting means for finding a third set of motion vectors from the collective error region to the selected pixel, and using the first set of motion vectors and the third set of motion vectors , A fourth to all pixels included in the current frame
A fourth motion vector generating means for generating a set of motion vectors, and a pixel value of each of the previous frames corresponding to one of the pixels of the current frame, through one of the motion vectors of the fourth set. Second identifying means for identifying the predicted current frame by assigning it as the one pixel value of the frame.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The video signal encoding apparatus and method of the present invention will be described in more detail below with reference to the drawings.

FIG. 1 shows a preferred embodiment of a video signal coding apparatus having a current frame prediction block according to the present invention. As shown in FIG. 1, the current frame signal is stored in the first frame memory 100.
The first frame memory 100 is connected to the subtractor 102 through the line L9 and is connected to the current frame prediction block 150 through the line L10.

In the current frame prediction block 150, the first
The current frame signal (line L10) extracted from the frame memory 100 and the reconstructed previous frame signal (line L12) from the second frame memory 124 are processed to predict the current frame on a pixel-by-pixel basis. The current frame signal is sent to the line L30, and the set (set) of motion vectors to the feature points is sent to L20. Details of the current frame prediction block 150 will be described with reference to FIG.

In the subtractor 102, the predicted current frame signal (line L30) is converted into the current frame signal (line L9).
And the resulting data, that is, the error signal representing the difference pixel value, is input to the video signal encoder 105. The error signal is encoded into a series of quantized transform coefficients using a discrete cosine transform (DCT) and quantization method. Next, the quantized transform coefficient is sent to the entropy encoder 107 and the video signal decoder 113. In the entropy encoder 107, the quantized transform coefficient from the video signal encoder 105 and the motion vector input via the line L20 from the current frame prediction block 150 use a variable length encoding technique. It is then encoded and sent to a transmitter (not shown) for its transmission.

On the other hand, in the video signal decoder 113, the quantized transform coefficient input from the video signal encoder 105 is transformed into a restored difference error signal by inverse quantization and inverse discrete cosine transform. To be done.

The restored error signal from the video signal decoder 113 and the predicted current frame signal on the line L30 from the current frame prediction block 150 are added to the adder 115.
Are added together to form a restored current frame signal, which is stored in the second frame memory 124 as the previous frame.

FIG. 2 shows a detailed view of the current frame prediction block 150 of FIG. The previous frame signal (line L12) from the first frame memory 124 of FIG.
It is input to the feature point selection block 210, the feature point motion vector detection block 212, the first motion compensation block 216, the support similar feature point and its motion compensation detection block 220.

In the feature point selection block 210, a plurality of feature points extracted from the pixels included in the previous frame are selected. Each of the feature points is defined as the position of a pixel that represents the motion of the object in the frame. In FIG. 6, 10
An exemplary frame of x7 pixels is shown. Here, if the moving object is closer to the center of the frame and the moving object is sufficiently represented by only one set of pixels, that is, "A" to "I" pixel values, these pixels are It is selected as a feature point of the frame.

In the preferred embodiment of the invention, FIG.
Feature points are identified using grid techniques using various forms of grids, such as the square grids or hexagonal grids shown in FIGS. As shown in the figure, those feature points are located at the nodes of the grid.

In another preferred embodiment of the present invention, an edge detection technique is used with the grid technique described above, as shown in FIGS. 8A and 8B. In this technique, the intersection of the grid and the edge of the moving object is selected as the feature point.

Referring again to FIG. 2, the selected feature points from the feature point selection block 210 are input to the feature point motion vector detection block 212, the first motion vector detection block 214 and the second motion vector detection block 222. To be done. Further, the feature point signal on the line L10 is supplied to the current frame motion vector detection block 212.

The feature point motion vector detection block 212 detects a first set of motion vectors for the selected feature points. Each of this first set of motion vectors is
It represents the spatial displacement between the feature point in the previous frame and the most similar pixel in the current frame.

There are various processing algorithms for detecting a motion vector on a pixel-by-pixel basis. In the preferred embodiment of the present invention, a block matching algorithm is used. That is, one feature point is the feature point selection block 210.
When received from, the feature point block having the feature point in the center of the feature point block, for example, the previous frame of 5 × 5 pixels is searched from the second frame memory 124 (see FIG. 1) through the line L12. Next, from the first frame memory 100 (see FIG. 1), the similarity between a feature point block and a plurality of candidate blocks of the same size included in a current search frame having a large size, for example, 10 × 10 pixels, is calculated. After the calculation, the motion vector of the feature point to the feature point block is specified.

After detecting the motion vectors to all the feature points, the first set of motion vectors is input to the first motion vector detection block 214 and the entropy coder 107 (see FIG. 1) via line L20. In the first motion vector detection block 214, a second set of motion vectors for all pixels included in the current frame is identified using the first set of motion vectors and the feature point information from the feature point selection block. It

In order to specify the second set of motion vectors, first, one set of motion vectors to "quasi-feature points" is specified, and each of these quasi-feature points is extracted from each feature point of the previous frame. It represents a pixel point of the current frame shifted by its corresponding first set of motion vectors. The magnitude of the motion vector to the quasi-feature point is equal to the magnitude of the motion vector to the corresponding feature point, but the directions of the two motion vectors are opposite. After specifying the motion vectors to all the quasi-feature points, the motion vectors to the non-quasi-feature points which are the remaining pixels in the current frame are specified as follows.

As shown in FIG. 9, a plurality of quasi-feature points are irregularly distributed in all the current frames. The motion vector to the non-quasi-feature point represented by a star is the radius (dr + d
It is obtained by averaging the motion vectors of the quasi-feature points located within the circle of a). Here, da is the distance between the most proximal quasi-feature point and the pixel position displayed by a star,
dr is a predetermined expansion radius including other quasi-feature points used in the calculation of the motion vector. For example, the most proximal feature point is “Y” and the feature point “X” is (dr
If it is included in the boundary of + da), the motion vector (MVx, MVy) to the pixel displayed by the star is calculated by the following equation.

[0027]

[Equation 1]

Here, dx and dy are the distances from the pixel positions displayed by stars to the quasi-feature points X and Y, respectively (M
Vx, MVy) X and (MVx, MVy) Y are motion vectors to the quasi-feature points, respectively.

Referring again to FIG. 2, the second set of motion vectors for the quasi-feature points and the non-quasi-feature points are provided to the first motion compensation block 216 via line L16. On the other hand, the part of the second set of motion vectors to the quasi-feature point is supplied to the support quasi-feature point and its motion vector detection block 220 via line L15. In the first motion compensation block 216, the second frame memory 124 (see FIG.
Each pixel value in the previous frame stored in the current frame as a value in one of the pixels in the current frame, and the pixel value is set to one of the pixels in the current frame through one in the second set of motion vectors. , And thus generate the preliminary predicted current frame. The output of the first motion compensation block 216 is the collective error region detection block 2
18 and supporting quasi-feature points and their motion vector detection block 220.

FIG. 3 is a block diagram of the collective error area detection block 218 shown in FIG. In the figure, the current frame is the line L10 from the first frame memory 100.
Through the first motion compensation block 216 to the subtractor 302 via line L17. The subtractor 302 is connected to an absolute value block 304. If the current frame signal is as shown in FIG. 10A and the preliminary predicted current frame signal is as shown in FIG. 10B, the outputs of subtractor 302 and extremum pairing block 304 are: It is represented as shown in FIG. 10C.

As shown in FIG. 10C, multiple error areas are found, for example, near the edges of the eyes, mouth and moving objects. The absolute value error signal from the absolute value block 304 is input to the filter 306. The filter 306 filters the absolute value error signal, and the fine error region is removed as shown in FIG. 10D. After being filtered, the filtered error signal is provided to the collective error region detection block 308, which is separated by a square window so that the collective error region signal as shown in FIG. It is transmitted onto L18.

Referring again to FIG. 2, the collective error region signal on line L18 is provided to the support quasi-feature point and its motion vector detection block 220, which is block 220.
Then, one pixel of each pixel of the collective error region is selected, and the motion vector of the selected pixel is specified as the support quasi-feature point and its motion vector. 4 and 5 are detailed block diagrams of the assisting quasi-feature point and its motion vector detection block 220 shown in FIG.

As shown in FIGS. 4 and 5, the collective error area signal is supplied to the motion vector selection block 400 and the motion vector detection block 402 to indicate the collective error area. This motion vector selection block 40
At 0, in response to the collective error region signal, the first group of motion vectors to the pixels in this collective error region are detected as the first motion vector detection block 214 (see FIG. 2).
From a second set of motion vectors provided by The motion vector of the first group is subtracted by the subtractor 406.
Supplied to

On the other hand, in the motion vector detection block 402, the second group of motion vectors to the pixels in the collective error area are specified between the current frame and the previous frame stored in the frame memory 404. The motion vector of the second group is also supplied to the subtractor 406.

Then, the subtractor 406 calculates the difference between the motion vectors of the two groups. The magnitude of the difference between each pixel is calculated in the absolute value block 408. The output of the absolute value block 408 is then divided by the magnitude of the difference between the two groups of motion vectors. If the magnitude of the difference is a value as shown in FIG. 11A, the collective error region is divided into two sub-blocks by the division block 410 as shown in FIG. 11B. FIG. 11B
In the situation as shown in, the pixel represented by a star becomes a supporting quasi-feature point of this collective error region, and its motion vector is selected as follows. First, a number of motion vector detection blocks 418 detect a sub-region motion vector for each of the two sub-regions by selecting a plurality of motion vectors for the sub-region, and then line this sub-region motion vector. The second motion vector detection / compensation block 422 is sent to the first motion vector detection / motion compensation block 420 and the switch SW1 via L42, and the second sub-region motion vector is sent to the second motion vector detection / motion compensation block 422 via line L44.
To the switch SW1. On the other hand, in the combination block 414, the center point of the collective error area selected from the center point selection block 412 is combined with the quasi-feature point supplied from the first motion vector detection block 214 through the line L15, and the result is obtained. Data of the first and second motion vector detection and motion compensation blocks 420, 42.
2 is supplied.

A first motion vector detection and motion compensation block 420 averages at least one of the motion vectors combined between the set of motion vectors to the quasi-feature points and the first subregion motion vector. Then, the motion vector for all pixels in the collective error region is searched for. Further, the first prediction region is specified by extracting each pixel value included in the collective error region from the second frame memory 124. The first prediction area is supplied to the first mean square error detection block 424.

A second motion vector detection and motion compensation block 422 averages at least one of the motion vectors combined between the set of motion vectors to the quasi-feature points and the second subregion motion vector. Then, the motion vector for all pixels in the collective error region is searched for. Also, the second prediction region is specified by searching each pixel value included in the collective error region from the second frame memory 124. This second prediction region is sent to the second mean square error detection block 426.

The first mean square error detection block 424 identifies the difference between the current frame and the first prediction region,
The result is sent to the compare and select block 428.
Similarly, in the second mean square error detection block 426,
The difference between the current frame and the second prediction region is identified and the result is sent to the compare and select block 428.

In this comparison and selection block 428, 2
The smaller of the outputs of the one mean square error detection blocks 424, 426 is identified and a switch control signal is provided to the switch SW1. This switch SW1 selects one of the two outputs on lines L42, L44,
Send to multiplexer 416. In the multiplexer 416, the output of the center point selection block 412 and the switch SW1
The motion vectors from are combined and the result is 1
It is supplied to the entropy encoder 107 as one supporting quasi-feature point and its motion vector.

Referring again to FIG. 2, in the second motion vector detection block 214, a third set of motion vectors for all pixels included in the current frame is obtained by using the quasi-feature points and the support quasi-feature points. Specified. Also, this third
The set motion vector is the second motion compensation block 224.
Supplied to In this second motion compensation block 224,
Each pixel value in the previous frame, corresponding to one of the pixels in the current frame through one of the third set of motion vectors stored in the second frame memory 124 (see FIG. 1), Assign as one value of pixel to generate the last predicted current frame.

The motion prediction block in the decoder corresponding to the encoder of the present invention is the feature point motion vector detection block 212, the first motion vector detection block 214, the first motion compensation block 216, and the collective error detection block 2.
18 has the same elements as those shown in FIG. This is because the first feature point motion vector, the supporting quasi-feature point and its motion vector are transmitted from and supplied to the encoder. Therefore, the motion prediction block comprises a feature point selection block having the same function as described for the encoder, a second current frame motion vector detection block and a second motion compensation block.

Further, the previous frame signal from the memory of the decoder is input to the feature point selection block and a plurality of feature points are selected. In the second current frame motion vector detection block, in response to the selected feature point and its motion vector and the supporting quasi feature point and its motion vector transmitted from the encoder described with reference to FIG. The motion vectors of all the pixels included in the predicted current frame are specified. The second motion compensation block provides the predicted current frame as at the encoder. This predicted current frame is further processed in the decoder to restore a current frame that is substantially the same as the original video signal.

[0043]

Therefore, according to the present invention, the image quality can be improved by encoding the video signal using the pixel-by-pixel motion prediction technique.

[Brief description of drawings]

FIG. 1 is a block diagram of a video signal encoding apparatus including a current frame prediction block of the present invention.

2 is a detailed block diagram of a current frame prediction block of FIG. 1. FIG.

FIG. 3 is a detailed block diagram of a collective error area detection block of FIG.

FIG. 4 is a detailed block diagram of a support quasi-feature point and its motion vector detection block of FIG. 2;

5 is a detailed block diagram of a support quasi-feature point and its motion vector detection block of FIG. 2;

FIG. 6 is a diagram illustrating an exemplary frame for defining a feature point.

FIG. 7 is a diagram of A and B, where A and B are examples of two types of grids used to select feature points.

FIG. 8 is a diagram of A and B, where A and B are techniques for selecting feature points through grids and edges, respectively.

FIG. 9 is a diagram showing a method for finding a motion vector to a non-quasi-feature point.

10A to 10E are diagrams showing a technique for searching a collective error region.

FIG. 11 is composed of A and B, and A and B are diagrams showing a technique for searching for a support quasi-feature point and its motion vector.

[Explanation of symbols]

100 First frame memory 102 Subtractor 105 Video signal encoder 107 Entropy encoder 113 Video signal decoder 115 Adder 124 Second frame memory 150 Current frame prediction block 210 Feature point selection block 212 Feature point motion vector detection block 214 first motion vector detection block 216 first motion compensation block 218 collective error region detection block 220 support quasi-feature point and its motion vector detection block 222 second motion vector detection block 224 second motion compensation block 302 subtractor 304 absolute value conversion Block 306 Filter 308 Collective error region detection block 400 Motion vector selection block 402 Motion vector detection block 404 Frame memory 406 Subtractor 408 Absolute value conversion block 41 Divided block 412 Center point selection block 414 Combination block 416 Multiplexer 418 Multiple motion vector detection block 420 First motion vector detection and motion compensation block 422 Second motion vector detection and motion compensation block 424 First mean square error detection block 426th 2 Mean Square Error Detection Block 428 Comparison and Selection Block

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Office reference number FI technical display location H04N 11/04 B 9185-5C

Claims (10)

[Claims] 1. A video signal encoding apparatus for use in a motion-compensated video signal encoder, which identifies a predicted current frame based on a current frame and a previous frame of a digital video signal. A first pixel selecting unit that selects a plurality of pixels from pixels included in the frame; and a first motion vector that represents a motion to each of the selected pixels between the current frame and the previous frame.
First motion vector selection means for finding a set of motion vectors, and a second motion for generating a second set of motion vectors for all pixels included in the current frame using the first set of motion vectors Assigning each pixel value in the previous frame corresponding to one of the pixels in the current frame as the one pixel value in the current frame through a vector generating means and one of the second set of motion vectors; A first specifying unit that specifies a preliminary predicted current frame; and a collective error region selecting unit that searches for a collective error region by specifying a difference between the current frame and the preliminary predicted current frame. And second pixel means for selecting one pixel from the pixels included in each of the collective error regions, and selecting from the collective error region A third motion vector selecting means for finding a third set of motion vectors for the selected pixels, and using the first set of motion vectors and the third set of motion vectors to select all the pixels included in the current frame. A fourth motion vector generating means for generating a fourth set of motion vectors, and a pixel value of each of the previous frames corresponding to one of the pixels of the current frame through one of the motion vectors of the fourth set. , A second specifying means for specifying the predicted current frame by assigning it as the one pixel value of the current frame. 2. The second motion vector generating means sets the motion vector of the first set to the pixel of the current frame in correspondence with the selected pixel of the previous frame to the pixel of the current frame of the second set. First allocating means for allocating as a part of the second set, and at least one motion vector included in a part of the second set of motion vectors is averaged to specify a motion vector of the remaining part of the second set of motion vectors. The video signal encoding device according to claim 1, further comprising a third specifying unit. 3. The collective error area selecting means subtracts the preliminary predicted current frame from the current frame to generate an error area signal, and an absolute value of the error area signal is calculated. Absolute value calculation means, filter means for filtering the absolute value error signal, windowing the filtered absolute value error signal, and window means for generating the collective error region signal. The video signal encoding device according to claim 1, further comprising: 4. The third motion vector selecting means divides each of the collective error regions into two or more sub-regions according to the size of each error pixel value included in each of the collective error regions. Dividing means, searching for sub-region motion vectors for each of the sub-regions, and generating sub-region motion vectors; sub-region motion vector generating means; and selecting one motion vector from the sub-region motion vectors, The video encoding device according to claim 1, further comprising a motion vector providing unit that provides each motion vector in the third set of motion vectors. 5. The fourth motion vector generating means, second allocating means for allocating the motion vector of the first set and the motion vector of the third set as a part of the motion vector of the fourth set, The video signal code according to claim 4, further comprising: fourth specifying means for averaging at least one motion vector included in a part of the four sets of motion vectors to specify a motion vector of the remaining part of the fourth set. Device. 6. Used in motion compensated video signal encoding,
A video signal encoding method for identifying a predicted current frame having a current frame and a previous frame of a digital video signal, comprising: (a) selecting a plurality of pixels from pixels included in the previous frame; A selection step, and (b) a first motion vector selection step of searching for a first set of motion vectors consisting of motion vectors representing a motion to each of the selected pixels between the current frame and the previous frame. (C) a second motion vector generation step of generating a second set of motion vectors for all pixels included in the current frame using the first set of motion vectors; and (d) the second Through one of the motion vectors in the set,
A first identifying step of identifying a preliminary predicted current frame by assigning each pixel value in the previous frame corresponding to one of the pixels in the current frame as the one pixel value in the current frame; (E) finding a collective error region by identifying a difference between the current frame and the preliminary predicted current frame;
Collective error area selection process, (f) 1 from the pixels included in each of the collective error areas
A pixel selection process of selecting two pixels, and (g) a third process from the collective error region to the selected pixel.
A third motion vector selection step of finding a set of motion vectors; and (h) using the first set of motion vectors and the third set of motion vectors, selecting a first motion vector for all pixels included in the current frame. A fourth motion vector generating process for generating four sets of motion vectors, and (i) through one of the fourth set of motion vectors,
A second identifying step of identifying the predicted current frame by assigning a pixel value of each of the previous frames corresponding to one of the pixels of the current frame as the one pixel value of the current frame; A video signal encoding method comprising: 7. The step (c) comprises: (c1) converting the first set of motion vectors to a pixel in the current frame corresponding to a selected pixel in the previous frame of the second set of motion vectors. (C2) averaging at least one motion vector included in a part of the second set of motion vectors, and averaging at least one motion vector of the remaining part of the second set of motion vectors; 7. The video signal encoding method according to claim 6, further comprising a third specifying step for specifying. 8. The step (e) includes: (e1) a subtraction step of subtracting the preliminary predicted current frame from the current frame to generate an error region signal; and (e2) generating the error region signal. An absolute value conversion step of converting to an absolute value; (e3) a filtering step of filtering the absoluteized error signal; and (e4) a windowing of the filtered absolute valued error signal to generate the collective error. 7. The video signal encoding method according to claim 6, further comprising a window process for generating a region signal. 9. The step (g) includes: (g1) dividing each of the collective error regions into two or more sub-regions according to the size of each error pixel value included in each of the collective error regions. A division process, (g2) a sub-region motion vector generation process for searching for a sub-region motion vector for each of the sub-regions, and generating a sub-region motion vector, and (g3) one motion vector from the sub-region motion vector And a motion vector providing step of providing each motion vector in the third set of motion vectors, the video signal encoding method according to claim 6. 10. The step (h) includes: (h1) a second allocation step of allocating the first set of motion vectors and the third set of motion vectors as a part of the fourth set of motion vectors; h2) averaging at least one motion vector included in a part of the fourth set of motion vectors to specify a motion vector of the remaining part of the fourth set. The video signal encoding method according to claim 9.