JP2002094947A - High-speed video playback device, high-speed video playback method, and recording medium - Google Patents

Abstract

(57) [Problem] To provide a high-speed video playback device, a high-speed video playback method, and a recording medium that can provide a more natural and smooth high-speed playback video with little loss of information even at high-speed playback. A video data storage unit for storing video data in digital format necessary for reproducing a video, and video data are sequentially read from the video data storage unit for each frame,
A video data reading unit that supplies the read video data to the first high-speed video generation unit 200, and N consecutive (N) images read from the video data storage unit via the video data reading unit.
≧ 2), a first high-speed video generation unit 200 having a first averaging unit 201 for generating a high-speed video frame for high-speed reproduction by displaying a high-speed video frame at a constant interval ΔT. And a video playback unit for playing back high-speed video.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a high-speed image reproducing apparatus,
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a high-speed video playback method and a recording medium on which a program for executing the method is recorded. Regarding the medium.

[0002]

2. Description of the Related Art Conventional high-speed video reproduction methods mainly include a VTR fast-forward reproduction method and a frame thinning-out reproduction method used when digital video data is fast-forwarded. First, a fast-forward playback method of a VTR will be described with reference to FIG. Reference numeral 2201 shown in FIG.
Is a tape-shaped storage medium, and an oblique band-shaped area denoted by reference numeral 2202 is a track storing video data, and a reference numeral 2203 denotes a reading head.

When a video recorded on a tape of a VTR is reproduced at a normal speed, a read head 2203 reads video data along a track 2202 as shown in FIG. An example of the video reproduced at this time is shown in FIG.
This is shown in FIG. On the other hand, when fast forward playback is performed,
As shown in (b), the reading head 2203 reads the video data so as to cross the plurality of tracks 2202 diagonally. The video reproduced at this time is shown in FIG.
As shown in (b), one screen is divided into strips, and each part of a plurality of video frames is displayed in each of the divided areas.

[0004] Next, with reference to FIG. 24, a method of thinning out and reproducing frames of a digital video will be described. In this method, high-speed playback is performed by thinning and displaying video frames in accordance with the playback speed. If the original video frames (F _{1 to} F ₉ ) shown in FIG. 24A are sequentially displayed at regular intervals, the video frames are reproduced at the normal speed. In addition, as shown in FIG. 24B, if only one frame is displayed every three frames, the reproduction is performed at triple speed.

[0005]

However, in the above-mentioned conventional high-speed image reproducing method, a problem may occur if there is a fast-moving object in the image. For example, in a fast-forward playback method of a VTR, a high-speed playback video is composed of a plurality of regions divided into bands, and each region displays a part of a different video frame. This object is displayed in a state where it is divided into strips.
At this time, if the movement of the object is slow, it is possible to recognize the divided object as one object although the movement is slightly different, but if the movement is fast, it is possible to recognize the divided object as one object, It is difficult to correctly recognize the shape of.

When a small object is included in an image, there is a possibility that the small object is not displayed because the high-speed playback image is composed of a plurality of divided areas. For example, in the example shown in FIG.
The clouds included in the second and third video frames of
It is not included in the high-speed playback video shown in (b). By performing the high-speed reproduction in this way, information necessary for the user may be lost.

[0007] On the other hand, in the digital video frame thinning reproduction method, since the thinned video frame is not displayed, an object having a short appearance time may not be displayed.
For example, if a video containing an object that appears only for less than 10 frames is played back at high speed at 10 times speed, the video frame containing this object will be skipped and the object will not be displayed at all. obtain. In addition, the moving distance of an object when a video including a fast-moving object is reproduced at a high speed is longer than the moving distance when the image is normally reproduced, so that it is difficult for a user to correctly recognize a fine moving path of the object.

[0008] Similarly, when the shape or color changes drastically, the above two high-speed reproduction methods may cause the high-speed reproduced video to be jerky or flickering.

As described above, since the original information is largely lost in the video reproduced at high speed by the conventional high-speed video reproduction method, the speed and size of the motion of the object included in the video,
Depending on the appearance time or the like, information required by the user is not reproduced, or even if reproduced, it is very difficult to recognize the information. Therefore, when a user visually searches for a scene while reproducing a video at a high speed, there is a possibility that a desired scene cannot be found.

The present invention has been made in view of the above-mentioned conventional problems, and has a high-speed video reproducing apparatus capable of providing a more natural and smooth high-speed reproduction video with little loss of information even at high-speed reproduction. It is an object of the present invention to provide a high-speed video reproduction method and a recording medium.

[0011]

According to a first aspect of the present invention, there is provided a high-speed video reproducing apparatus which averages a plurality of continuous video frames constituting a video,
A simple averaging unit for generating a high-speed video frame for high-speed reproduction, and a video reproducing unit for displaying the high-speed video frame generated by the simple averaging unit at regular intervals.

According to a second aspect of the present invention, there is provided a high-speed video frame for high-speed playback, wherein each of a plurality of continuous video frames constituting a video is weighted to average the plurality of video frames. And a video reproducing means for displaying the high-speed video frames generated by the weighted averaging means at regular intervals.

According to a third aspect of the present invention, there is provided a high-speed video reproducing apparatus, comprising: a video frame dividing means for dividing a plurality of continuous video frames constituting a video into a plurality of regions; An area change detecting means for detecting a degree of change over the plurality of video frames with respect to an area; and an area having a degree of change detected by the area change detecting means of not less than a predetermined level. Averaging means for performing weighted averaging with averaging or weighting, video data of the area averaged or weighted averaged by the averaging means, and the degree of change detected by the area change detection means Video data synthesis for generating high-speed video frames for high-speed playback by synthesizing video data of a predetermined video frame in an area smaller than a predetermined level And the step, the is a high-speed video frame generated by video data combining means that a video reproduction means for displaying at regular intervals.

According to a fourth aspect of the present invention, in the high-speed image reproducing apparatus according to the third aspect, the area change detecting means determines a degree of change based on an average of hue or luminance of the image frame. It is to detect.

According to a fifth aspect of the present invention, there is provided a high-speed image reproducing apparatus according to the first, second, third or fourth aspect, wherein the simple averaging means, the weighted averaging means or the averaging means is provided. The inverse orthogonal transform is performed by averaging or weighted averaging a plurality of orthogonally coded video frames to generate a high-speed video frame represented by a transform coefficient, and performing inverse orthogonal transform on the high-speed video frame represented by the transform coefficient. It further comprises orthogonal transform means.

According to a sixth aspect of the present invention, in the high-speed image reproducing apparatus according to the first, second, third, fourth or fifth aspect, when the image is reproduced at a high speed at N times speed, the simple averaging is performed. The means, the weighted averaging means, or the averaging means averages or weights the N video frames to generate one high-speed video frame.

According to a seventh aspect of the present invention, there is provided a high-speed video playback apparatus according to the first, second, third, fourth or fifth aspect of the present invention. When there is a scene break, the simple averaging means, the weighted averaging means or the video frame selecting means for selecting the video frame to be averaged or weighted averaged by the averaging means according to the break position. is there.

The high-speed video reproducing apparatus according to claim 8 is the high-speed video reproducing apparatus according to claim 7, further comprising a break detecting means for detecting a break of a scene in the predetermined number of video frames. is there.

A high-speed video reproducing method according to claim 9, wherein a simple averaging step of averaging a plurality of continuous video frames constituting a video to generate a high-speed video frame for high-speed reproduction, A video playback step of displaying the high-speed video frames generated in the conversion step at regular intervals.

According to a tenth aspect of the present invention, there is provided a high-speed video frame for high-speed playback, wherein each of a plurality of continuous video frames constituting a video is weighted to average the plurality of video frames. And a video reproduction step of displaying the high-speed video frames generated in the weighted averaging step at regular intervals.

Further, in the high-speed video reproducing method according to claim 11, a plurality of continuous video frames constituting a video are divided into a plurality of regions, and each of the plurality of video frames divided in the video frame dividing step is divided. An area change detecting step of detecting a degree of change over the plurality of video frames with respect to an area; and an area in which the degree of change detected in the area change detecting step is equal to or greater than a predetermined level. Averaging step of performing weighted averaging with averaging or weighting, video data of the area averaged or weighted averaged in the averaging step, and the degree of change detected in the area change detection step is A video that combines high-speed video frames for high-speed playback by combining video data of a predetermined video frame in an area smaller than a predetermined level. And data combining step is a high-speed video frame said generated by the video data synthesis step one having a video playback step of displaying at regular intervals.

According to a twelfth aspect of the present invention, in the high-speed image reproducing method according to the eleventh aspect, the area change detecting step determines a degree of change based on an average of hue or luminance of the image frame. It is to detect.

According to a thirteenth aspect of the present invention, in the high-speed image reproducing method according to the ninth, tenth, eleventh or twelfth aspect, the simple averaging step, the weighted averaging step or the averaging step is performed. The inverse orthogonal transform is performed by averaging or weighted averaging a plurality of orthogonally coded video frames to generate a high-speed video frame represented by a transform coefficient, and performing inverse orthogonal transform on the high-speed video frame represented by the transform coefficient. It further comprises an orthogonal transformation step.

According to a fourteenth aspect of the present invention, in the high speed video reproducing method according to the ninth, tenth, eleventh, or thirteenth aspect, when the video is reproduced at a high speed at N times speed, the simple averaging is performed. The step, the weighted averaging step, or the averaging step is for averaging or weighted averaging the N video frames to generate one high-speed video frame.

According to a fifteenth aspect of the present invention, there is provided the high-speed video reproducing method according to the ninth, tenth, eleventh, twelfth or thirteenth aspect. When there is a scene break,
The method may further include a video frame selecting step of selecting a video frame to be averaged or weighted averaged by the simple averaging step, the weighted averaging step, or the averaging step according to a break position.

A high-speed video playback method according to a sixteenth aspect of the present invention is the high-speed video playback method according to the fifteenth aspect, further comprising a break detection step of detecting a break of a scene in the predetermined number of video frames. .

Further, a computer-readable recording medium according to claim 17 is a computer-readable recording medium.
It is recorded as a program for causing a computer to execute the high-speed video reproduction method described in 1, 12, 13, 14, 15, or 16.

[0028] A high-speed video playback apparatus according to claim 1 of the present invention, a high-speed video generation method according to claim 9, and claim 17.
In the recording medium according to the first aspect, the simple averaging means (simple averaging step) averages a plurality of continuous video frames constituting the video to generate a high-speed video frame for high-speed playback, and the video playback means (the video playback step 3), high-speed video frames are displayed at regular intervals.

Therefore, even if the image is reproduced at a high speed, the information of all the image frames is reflected without any loss of information due to the thinning of the image frame, so that the image almost similar to the original image can be reproduced smoothly and naturally at a high speed. In addition, even if the video contains fast moving objects, small objects, objects with a short appearance time, etc., the high speed video is reproduced by displaying the high speed video frame which averaged the video frame. The object is always reproduced. For this reason, a user who visually searches for a video can easily search for a desired scene at the same playback speed as compared with the related art, and can obtain a desired scene with the same accuracy and faster. , It is possible to search in a short time.

Also, a high-speed video reproducing apparatus according to claim 2,
In the high-speed video generation method according to the tenth aspect and the recording medium according to the seventeenth aspect, the weighted averaging means (weighted averaging step) assigns weights to each of a plurality of continuous video frames constituting the video, and Are averaged to generate high-speed video frames for high-speed playback, and the high-speed video frames are displayed at regular intervals in the video playback means (video playback step).

In the high-speed video frame generated by the simple averaging, all video frames are displayed only with uniform clarity, and the user who searches the high-speed video visually recognizes particularly fast moving objects, small objects, Although it is not easy to find an object with a short appearance time, the object included in the video frame with a large weight is clearly displayed, so that the user can easily find a desired object included in the video. Can be found.

Also, a high-speed video reproducing apparatus according to claim 3,
In the high-speed video generation method according to the eleventh aspect and the recording medium according to the seventeenth aspect, the video frame dividing means (video frame dividing step) divides a plurality of continuous video frames constituting the video into a plurality of regions, In the area change detecting means (area change detecting step), the degree of change over a plurality of video frames is detected for each area, and in the averaging means (averaging step), the area change detecting means (area change detecting step) is used. For a region where the detected degree of change is equal to or greater than a predetermined level, average or weighted averaging of a plurality of video frames is performed,
In the video data synthesizing means (video data synthesizing step), the video data of the area averaged or weighted by the averaging means (averaging step) and the change detected by the area change detecting means (area change detecting step) A high-speed video frame for high-speed playback is generated by synthesizing the video data of a predetermined video frame in a region where the degree of the video signal is smaller than a predetermined level, and the high-speed video frame is displayed at regular intervals in the video playback means (video playback step) are doing.

In an area where the degree of change synthesized by the image data synthesizing means (image data synthesizing step) is smaller than a predetermined level, the image data of one image frame is used for the image data of another image frame. Although the area is thinned out, the amount of information missing due to the thinning is small because the change in this area is small. Further, video data used in an area with a small change included in a high-speed video frame is not subjected to processing such as averaging, so that the video in that area can be sharpened. On the other hand, for an area in which the degree of change is equal to or greater than a predetermined level, an averaged video obtained by averaging or weighted averaging of a plurality of video frames is reproduced. High-speed video can be played.

In particular, a high-speed video reproducing apparatus according to claim 4,
In the high-speed image generation method according to the twelfth aspect and the recording medium according to the seventeenth aspect, the area change detection unit (area change detection step) may detect the degree of change based on the average of the hue or luminance of the video frame. desirable. At this time, the area change detecting means (area change detecting step) measures the average value of the hue or luminance of the area, and compares the amount of change over a plurality of video frames with a predetermined level.

A high-speed video reproducing apparatus according to claim 5,
In the high-speed image generation method according to claim 13 and the recording medium according to claim 17, the simple averaging means (simple averaging step), the weighted averaging means (weighted averaging step) or the averaging means (averaging step) Averaging or weighted averaging of a plurality of orthogonally coded video frames to generate a high-speed video frame represented by a transform coefficient, which is represented by a transform coefficient in an inverse orthogonal transform means (inverse orthogonal transform step). High-speed video frames are subjected to inverse orthogonal transform.

Therefore, even if a video frame is orthogonally coded, the average is directly averaged by the simple averaging means (simple averaging step), the weighted averaging means (weighted averaging step) or the averaging means (averaging step). Therefore, a high-speed video frame can be generated in a short time without having to perform inverse conversion of all the plurality of video frames.

A high-speed video reproducing apparatus according to claim 6,
In the high-speed video generation method according to the fourteenth aspect and the recording medium according to the seventeenth aspect, when the video is played back at N times speed,
The simple averaging means (simple averaging step), the weighted averaging means (weighted averaging step) or the averaging means (averaging step) averages or weights the N video frames to obtain one high-speed video. It is desirable to generate frames.

Also, a high-speed video reproducing apparatus according to claim 7,
In the high-speed video generation method according to the fifteenth aspect and the recording medium according to the seventeenth aspect, when there is a scene break in the middle of a predetermined number of continuous video frames constituting a video, a video frame selection unit (video frame selection step) , The simple averaging means (simple averaging step), the weighted averaging means (weighted averaging step), or the averaging means (averaging step) selects an image frame to be averaged or weighted averaged according to the break position. ing.

Therefore, by averaging or weighted averaging only video frames of the same scene, and generating a high-speed video frame, a plurality of scenes are not mixed. Can be. As a result, even when there are many scenes, a user who searches for a high-speed video visually can easily perform the search.

Furthermore, a high-speed video reproducing apparatus according to claim 8, a high-speed video generating method according to claim 16, and claim 1.
In the recording medium according to the seventh aspect, it is preferable that a break detecting means (a break detecting step) detect a break of a scene in a predetermined number of video frames.

[0041]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An image displayed on a display of a television or a personal computer is reproduced by displaying a still image (video frame) at a constant interval of about 30 frames per second, that is, by refreshing. However, since the frame rate (the number of video frames displayed per second) is the same as that during normal playback even when the video is played back at high speed, when the video is played back at N times speed, N out of N video frames are used. One piece of information is missing. In order to reproduce a video at high speed without loss of information, video frames must be displayed at N times the frame rate. That is, normally, one video frame is displayed every 33 msec.
N video frames must be displayed every second.

However, even if N video frames are displayed during a short time of 33 msec, human eyes cannot refresh these N frames because the refresh is too fast.
A video frame is not captured as a moving image but captured as a still image. That is, human eyes see the N video frames as an averaged still image. In the high-speed video playback device and high-speed video playback method according to the present invention described below, when video is played back at N times high speed, N video frames are averaged to generate a high-speed video frame, and a high-speed video frame is generated at a specified frame rate. This high-speed video frame is displayed.

The first embodiment, the second embodiment, the third embodiment, and the fourth embodiment of the high-speed video reproducing apparatus and the high-speed video reproducing method according to the present invention will be described below. Embodiment] will be described in detail with reference to the drawings.
In the description of each embodiment, the high-speed video playback device and the high-speed video playback method according to the present invention will be described in detail, but a program for executing the high-speed video playback method is recorded on the recording medium according to the present invention. Since it is a recording medium, its description is included in the following description of the high-speed video reproduction method.

[First Embodiment] FIG. 1 shows a first embodiment of the present invention.
FIG. 2 is a block diagram showing a high-speed video playback device according to the embodiment. In FIG. 1, a high-speed video playback device according to the present embodiment includes a video data storage unit 101 and a video data reading unit 1.
03, a first high-speed video generation unit 200, and a video reproduction unit 105 corresponding to the video reproduction means in the claims.

First, the video data storage unit 101 stores digital video data required for reproducing a video, and includes a hard disk (HD) and a DVD (Division).
gital Versatile Disc). The format of the video data stored in the video data storage unit 101 is a format in which frame data in RGB format is continuous. The video data reading unit 10
Reference numeral 3 denotes a unit for sequentially reading video data from the video data storage unit 101 for each frame, and supplying the read video data to the first high-speed video generation unit 200 for each video frame.

Further, the first high-speed video generation section 200 transmits the video data to the video data storage section 101 via the video data reading section 103.
Averaging N (N ≧ 2) consecutive video frames read out from the multiplexing section to generate a high-speed video frame for high-speed reproduction. Further, the video reproduction unit 105 reproduces the high-speed video by displaying the high-speed video frames generated by the first high-speed video generation unit 200 at a constant interval ΔT.
When a video is reproduced at a normal reproduction speed (1 × speed) without performing high-speed reproduction, the video frames read from the video data storage unit 101 are provided to the video reproduction unit 105 without averaging, It is displayed at regular intervals ΔT.

The internal structure and operation of the first high-speed video generator 200 will be described below with reference to the first embodiment and the second embodiment.
Two cases will be described. (First Embodiment) First, in the first embodiment, the first high-speed video generation unit 200 internally includes a first averaging unit 201 corresponding to a simple averaging unit according to the claims as shown in FIG.
It is provided with. The video data storage unit 10
The video data read from 1 is usually discrete cosine transform (DCT), quantization and variable length coding (Variable Length Coder: VL).
C), the video data supplied to the first averaging unit 201 of the present embodiment is assumed to be pure video data that has been subjected to variable-length decoding, inverse quantization, and inverse DCT.

The first averaging section 201 averages N (N ≧ 2) video frames read from the video data storage section 101 to generate a high-speed video frame. For example, the video frames F _{1 to} F ₃ shown in FIG.
Is tripled, the first averaging unit 201 obtains the high-speed video frame S shown in FIG. 3B by superimposing and averaging the video frames F _{1 to} F ₃ . The first averaging unit 201 continuously outputs high-speed video frames S by averaging three consecutive video frames F, as shown in FIG.

Referring now to FIG. 5, first averaging section 2
01 will be described in detail. Each of the video frames F _{1 to} F ₃ is represented by 4 × 4 pixels, and the pixel value P of the coordinates (x, y) in the color C (any of RGB) of the video frame Fi is represented by P
When represented by (Fi, C, x, y), the pixel value Ps of the high-speed video frame S for triple-speed playback is obtained by the following equation (1).

[0050]

(Equation 1)

For example, the upper left pixel P (S, C, 1,1) of the high-speed video frame S is represented by P (S, C, 1,1) =
(180 + 180 + 180) / 3 = 180, and the pixel P (S, S) of the high-speed video frame S at the coordinates (1,3)
C, 1,1) is P (S, C, 1,3) = (0 + 180)
+180) / 3 = 120.

As described above, the pixel values at the coordinates (x, y) of the high-speed video frame S for N-times speed reproduction generated by the first averaging unit 201 using the video frames F _{M + 1 to} F _{M + N.} Ps is obtained by the following equation (2).

[0053]

(Equation 2)

The first averaging unit 201 generates a high-speed video frame S by calculating pixel values Ps for all colors (RGB) and all coordinates using the above equation (2). Send it to playback section 105.

(Second Embodiment) Next, as a second embodiment,
The internal configuration of the first high-speed video generator 200 will be described. Note that the video data stored in the video data storage unit 101 is video data that has been subjected to DCT, quantization, and variable length coding. In the present embodiment, the first high-speed video generation unit 200 internally includes a variable length decoding (VLD) unit 203, an inverse quantization unit 205, and a simple average in claims as shown in FIG. A second averaging unit 207 corresponding to the averaging unit and an inverse DCT unit 209 corresponding to the inverse orthogonal transform unit in the claims are provided.

First, the VLD unit 203 performs variable length decoding of the video data read from the video data storage unit 101, and performs variable length decoding for each video frame.
If the user requests N-times high-speed playback, N
N VLD units 203 are prepared so that variable-length decoding can be performed on one video frame in parallel. Also,
The inverse quantization unit 205 inversely quantizes the video frames that have been subjected to the variable length decoding in the VLD unit 203, and, like the VLD unit 203, prepares N pieces of inverse quantization units so that they can be inversely quantized in parallel for each video frame. ing.

Further, the second averaging section 207 performs N (N ≧ N) represented by DCT coefficients that have been subjected to variable-length decoding and inverse quantization by the VLD section 203 and the inverse quantization section 205.
The 2) video frames are averaged to generate a high-speed video frame represented by a DCT coefficient (hereinafter, referred to as a DCT high-speed video frame).

The second averaging section 207 of the present embodiment performs
The DCT coefficient of each video frame is averaged by an averaging method similar to the averaging unit 201 of FIG. That is, DCT-processed video frames (hereinafter referred to as DCT video frames) F _D1 to
If each of the F _D3 consists DCT coefficients of 4 × 4 as shown in FIG. 7, the user has requested a fast reproduction of triple speed, the second
The averaging unit 207 calculates the upper left DCT coefficient, which is the first frequency component of the DCT high-speed video frame _SD , by (10
0 + 180 + 250) / 3 ≒ 177.

Further, the inverse DCT section 209 converts the DCT high-speed video frame generated by the second averaging section 207 into an inverse D
The CT is used as a high-speed video frame that can be reproduced by the video reproduction unit 105, and the high-speed video frame is sent to the video reproduction unit 105.

Next, the operation of the high-speed video reproducing apparatus of this embodiment for averaging video frames and reproducing the video at high speed will be described with reference to FIG. First, FIG.
Are displayed at the normal reproduction speed, the video frames Fa to Fg are displayed every 33 msec from the video frame Fa in order. At this time, as shown in FIG. 8B, an image in which the ball bounces once and lands is reproduced.

For reference, if triple-speed playback is performed by a conventional method of performing high-speed playback by a frame thinning playback method in which video frames are thinned out and displayed, as shown in FIG. 8C, video frames Fa, Fd, Since Fg is displayed in order, to the human eye, as shown in FIG. 8D, the ball does not bounce and appears to fall straight on toward the ground mark. In addition, since there is a distance between the balls between the video frames Fa, Fd, and Fg, the balls may be recognized as a plurality of balls and may not appear to be moving.

On the other hand, the high-speed video reproducing method of the present embodiment is used.
And 3 × speed reproduction, video frames Fa to Fc, F
d to Ff and Fg are respectively averaged and shown in FIG.
Three high-speed video frames S₁~ S_ThreeIs generated
You. Three high-speed video frames S ₁~ S_ThreeIs a fast video frame
Room S₁Is displayed every 33 ms in order from
Eyes are 3 times faster and the ball travel path is
The video frame Fa shown in FIG.
~ Fg looks almost the same as the video displayed every 11ms
Become.

As described above, according to the high-speed video reproducing apparatus and the high-speed video reproducing method of the present embodiment, when reproducing a video frame at a high speed, a plurality of video frames need not be thinned out as in the related art. By averaging the video frames, high-speed video frames are generated and displayed. In this way, even in high-speed playback, the information of all video frames is reflected without loss of information due to thinning,
A video almost similar to the original video can be reproduced at a high speed in a smooth and natural form.

Even if the video includes a fast moving object, a small object, an object with a short appearance time, etc., the high speed video is reproduced by displaying the high speed video frame obtained by averaging the video frames. These objects are always reproduced. Therefore, a user who visually searches for a video can easily search for a desired scene at the same playback speed as compared with the related art, and can search for a desired scene with the same accuracy and faster. Because it can be found, the search can be performed in a short time.

Further, according to the high-speed video reproducing apparatus provided with the first high-speed video generating section 200 described in the second embodiment, the video data stored in the video data storing section 101 is the data obtained by performing the discrete cosine transform. However, since the second averaging unit 207 can perform averaging as it is, there is no need to perform inverse DCT on video data before averaging, and a high-speed video frame can be generated in a short time.

In this embodiment, the format of the video data stored in the video data storage unit 101 is R
Although the GB format is used, any format may be used. For example, a compressed video format such as Motion JPEG or a YUV format may be used.

[Second Embodiment] FIG. 9 shows a second embodiment of the present invention.
FIG. 2 is a block diagram showing a high-speed video playback device according to the embodiment. In the figure, the same reference numerals are given to the same parts as those in FIG. 1 (first embodiment), and the description is omitted. The high-speed video playback device according to the present embodiment includes a second high-speed video generation unit 300 instead of the first high-speed video generation unit 200 included in the high-speed video playback device according to the first embodiment.

The second high-speed video generation unit 300 performs weighted averaging of N consecutive (N ≧ 2) video frames read from the video data storage unit 101 via the video data reading unit 103, and performs high-speed reproduction. As shown in FIG. 10, a weighted averaging unit 301 corresponding to a weighted averaging unit in claims is provided.

The weighted averaging section 301 weights and averages video frames to generate high-speed video frames, instead of simple averaging as performed by the first averaging section 201 of the first embodiment. Things. For example, when the 3-speed playback video frames F ₁ to F ₃ shown in FIG. 11 (a), the weighted average unit 301, by weighted averaging the video frame F ₁ to F _3, FIG. 11 (b ) Is obtained. The video frames F _{1 to} F ₃ are displayed in the order of the video frames F ₁ , F ₂ , and F _{3 in} the case of normal reproduction. In the present embodiment, the weight is increased in the temporally later video frame. fast video frame Sw generated by the averaging unit 301 is assumed to reflect strongly the video frame F _3.

The weighted averaging section 301 internally has a multiplication section 303 for performing the weighted averaging as described above.
And an adder 305 and a divider 307. The multiplication unit 303 assigns different weights to the video data read from the video data storage unit 101 for each video frame, and multiplies all pixel values of the video frame by a predetermined weighting coefficient. In the present embodiment, as described above, the weight increases with time in a later video frame. That is, of the video frames to be averaged, the first video frame is multiplied by the weighting factor 1, the next video frame is multiplied by the weighting factor 2, and the N-th video frame is weighted by the weighting factor N Multiply by Note that the multiplication unit 3
A plurality of reference numerals 03 are provided so as to multiply video frames by different weight coefficients.

The adder 305 adds a plurality of video frames multiplied by different weighting factors. Further, the division unit 307 includes the addition unit 30
The high-speed video frame is output by dividing the video frame added by 5 by the cumulative number M of the used weighting factors. Note that, when 1 to N are used as the weight coefficients, the cumulative number M of the weight coefficients is M = 1 + 2 +... + N = N
(1 + N) / 2.

Hereinafter, referring to FIG. 12, weighted averaging section 3
01 will be described in detail. Each of the video frames F _{1 to} F ₃ is 4 pixels × 4
The color C (RG) of the video frame Fi is expressed in pixels.
B), the pixel value P at the coordinates (x, y)
Is represented by P (Fi, C, x, y), the pixel value P of the high-speed video frame Sw for triple-speed playback in the present embodiment is represented by P (Fi, C, x, y).
s is obtained by the following equation (3).

[0073]

(Equation 3)

For example, the upper left pixel P (S, C, 1,1) of the high-speed video frame Sw is P (S, C, 1,1)
= (1 × 180 + 2 × 180 + 3 × 180) / 6 = 18
0, and the pixel P (S, C, 1,4) of the high-speed video frame Sw at the coordinates (1,4) is P (S, C, 1,4)
= (1 × 0 + 2 × 0 + 3 × 180) / 6 = 90.

As described above, the pixel value Ps at the coordinates (x, y) of the high-speed video frame Sw for N-times speed reproduction generated by the weighted averaging unit 301 using the video frames F _{M + 1 to} F _{M + N} is , Is obtained by the following equation (4).

[0076]

(Equation 4)

The weighted averaging section 301 calculates the above equation (4)
Is used to calculate a pixel value Ps for all colors (RGB) and all coordinates, thereby generating a high-speed video frame Sw and sending it to the video reproduction unit 105.

As described above, when the video frames are reproduced at high speed in the present embodiment, weighted averaging with weighting is performed instead of simply averaging the video frames as in the first embodiment. Generates a high-speed video frame and displays it. Therefore, a video frame with a larger weight is strongly reflected in a high-speed video frame, and is reproduced more clearly.

For example, as shown in FIG. 13, high-speed video frames (FIG. 1) obtained by averaging in the first embodiment
3 (b)), since the density of each object (triangle) included in the original video frame (FIG. 13 (a)) is all uniform, the object melts into the background. Although the portion is not clearly displayed, in the high-speed video frame obtained by the weighted averaging of the present embodiment (FIG. 13C), the object included in the video frame multiplied by the large weight coefficient is replaced by another video frame. It is displayed darker and clearer than the included objects. Also, an object that is displayed for a short time is displayed more clearly than a high-speed video frame generated by simply averaging.

As described above, according to the high-speed video reproducing apparatus and the high-speed video reproducing method of the present embodiment, when averaging video frames, weighted averaging using weighting factors is performed instead of uniform averaging. Is going. For this reason, the high-speed video reproduced by using the high-speed video frame generated by the weighted averaging is different from the high-speed video generated in the first embodiment in the video frame multiplied by a large weighting factor. Objects are displayed more clearly. Therefore, a user who visually searches a high-speed image can easily find a fast-moving object, a small object, or an object with a short appearance time even if the image includes the object.

In this embodiment, a larger weighting factor is applied to a temporally later video frame, but a larger weighting factor may be applied to a temporally earlier video frame. Further, in the present embodiment, the weighting coefficient is a positive integer (1,
2,..., N), but may be a decimal number. Also,
A video frame including many complex high-frequency components with a texture may be multiplied by a large weighting factor. At this time, details appear clearly in the high-speed video frame. Further, the weight coefficient may be increased for a video frame including many predetermined colors. At this time, a user who visually searches for an image based on a color can easily search for an object of a predetermined color existing in the image in a short time.

Also in this embodiment, as in the second example of the first embodiment, if the video data has been subjected to discrete cosine transform, the weighted averaging unit 301 causes the DCT video represented by the DCT coefficient to The above-described weighted averaging may be performed on the frame.

[Third Embodiment] FIG. 14 is a block diagram showing a high-speed video reproducing apparatus according to a third embodiment of the present invention. In the figure, the same reference numerals are given to the same parts as those in FIG. 1 (first embodiment), and the description is omitted.
The high-speed video playback device according to the present embodiment is configured to include a third high-speed video generation unit 400 instead of the first high-speed video generation unit 200 included in the high-speed video playback device according to the first embodiment.

The third high-speed video generation unit 400 divides each of the N consecutive video frames read from the video data storage unit 101 via the video data reading unit 103 into a plurality of regions, and As a result of detecting the change, the image frame is thinned out in the region where the change is small, and the averaging is performed in the region where the change is large. As shown in FIG. 15, the third high-speed video generation unit 400 includes a video frame division unit 401 corresponding to a video frame division unit described in claims, an area change detection unit 403 corresponding to a region change detection unit, A third averaging unit 405 corresponding to the averaging unit and a video data combining unit 407 corresponding to the video data combining unit are provided.

First, the video frame division unit 401 divides each of N (N ≧ 2) video frames read from the video data storage unit 101 into a plurality of areas. The area change detection unit 403 detects the degree of change over N video frames for each of the areas divided by the video frame division unit 401. Are transmitted to the video data synthesizing unit 407.

The degree of the change is detected based on the average value of the hue and the luminance in the area. For example, when the luminance changes greatly between temporally adjacent video frames, the area change detecting unit 403 detects the change. It is determined that the image in the area has changed significantly. Note that the degree of change may be determined from a total value obtained by accumulating differences in hue and luminance between temporally adjacent video frames over N video frames. At this time, the area change detection unit 403 determines that there is a large change in the area where the total value exceeds the specified value.

The third averaging section 405 averages the area determined to have a large change by the area change detecting section 403 over N video frames, and generates the high-speed reproduction generated by the averaging. Are supplied to the video data synthesizing unit 407. In the present embodiment, the third averaging unit 405 is described as performing the simple averaging described in the first embodiment. However, the third averaging unit 405 performs weighting using the weight coefficient described in the second embodiment. Averaging may be performed. Also, as in the second example of the first embodiment, if the video data has been subjected to the discrete cosine transform, the third averaging unit 405 performs averaging on the DCT video frame represented by the DCT coefficient. Alternatively, weighted averaging may be performed.

Further, the video data synthesizing section 407 synthesizes the video data from the third averaging section 405 and the video data of the small change area sent from the area change detection section 403 to reproduce the data at high speed. To generate a high-speed video frame for use. When synthesizing the video data, as the video data of the area where the change is small, of the N video frames,
Only the video data of the corresponding area of the first video frame is synthesized, and the video data of the corresponding area of another video frame is not used by thinning. Also,
In the present embodiment, it is assumed that the video data of the first video frame is used as the video data of the small change area for synthesis, but it is considered that the video data of every video frame is almost the same in the small change area. Therefore, video data of any video frame may be used.

Next, an operation for generating a high-speed video frame for high-speed video reproduction by the high-speed video reproduction device of the present embodiment will be described with reference to FIG. First,
The four video frames F _{1 to} F ₄ shown in FIG.
It is divided into six regions. These four video frames F ₁
When playing to F _4, the object in the upper region (ball) is while moving from the left to the right, almost no change appears lined block image is in the lower region. For this reason,
The area change detection unit 403 determines that the eight areas of the upper half (2 × 4) are the parts having a large change, sends the video data of these areas to the third averaging unit 405, and sends the remaining lower half ( 2
The eight areas of (× 4) are determined to be small changes, and the image data of these areas is sent to the image data synthesizing unit 407.

The third averaging section 405 averages the upper half video data of each of the video frames F _{1 to} F ₄ sent from the area change detection section 403 for each area, and outputs the averaged video data. The data is sent to the video data synthesizing unit 407. next,
The video data combining unit 407 combines the video data sent from the third averaging unit 405 with the video data of a small change sent from the area change detecting unit 403, as shown in FIG. A high-speed video frame for quadruple-speed playback shown is generated.

As described above, in the high-speed video reproducing apparatus and the high-speed video reproducing method according to the present embodiment, in the generation of the high-speed video frame, the video of one video frame is reproduced in the area where the change is small. Although video data of other video frames is thinned out, the amount of information missing due to the thinning is small because the change is small. In addition, since processing such as averaging is not performed on video data of a small change area included in the high-speed video frame, the video of the area can be sharpened. On the other hand, for an area having a large change, a video in which a plurality of video frames are averaged is reproduced as in the first embodiment, so that a smooth and natural high-speed video without information loss due to thinning is reproduced. can do.

When generating a high-speed video frame without dividing the video frame into a plurality of regions, the generation of a high-speed video frame by thinning out N video frames as in the prior art is performed. Generation of high-speed video frames by simple averaging as in the embodiment or weighted averaging using weighting coefficients as in the second embodiment, or generation of high-speed video frames by combining thinning and averaging as in this embodiment A generation type selection unit (not shown) for selecting one of the generations depending on the degree of change of the N video frames may be further provided. By doing so, the first technology according to the prior art and the present invention can be changed according to the degree of change of the video frame.
The third embodiment can be selectively executed by switching between the third embodiment and the third embodiment.

[Fourth Embodiment] The high-speed video reproducing apparatus and the high-speed video reproducing method according to the first to third embodiments generate one high-speed video frame from N video frames. In the high-speed video playback device and the high-speed video method according to the embodiment, the number of frames is not limited to N.
A high-speed video frame may be generated from around N video frames in accordance with the cut point. For example, if is shown that there is a cut point B between the image frame F ₄ and the video frame F ₅ in the video frame F ₁ to F ₉ shown in FIG. 17 (a). Hereinafter, a scene comprising a scene consisting of a video frame F ₁ to F ₄ Scene 1, the image frame F ₅ to F ₉ scene 2.

At this time, the high-speed video reproduction of the first embodiment is performed.
Device for triple-speed playback using the apparatus and high-speed video playback method
The fast video frame is, as shown in FIG.
Laem F₁~ F_Three, Video frame F_Four~ F₆, Picture frame
F₇~ F₉High-speed video frame S generated from each of
₁, S_Two, S_ThreeBecomes However, high-speed video frames
S_TwoIs a frame in which scene 1 and scene 2 are mixed
Information that humans can recognize.
A stool, especially an object included in the video frame F4
Recognition of (ball) is difficult. For this reason, the present embodiment
In the high-speed video reproducing apparatus and the high-speed video reproducing method of FIG.
As shown in FIG. 7 (b), the video frame F ₁~ F_Four,
Laem F_Five, F₆, Video frame F₇~ F₉From each of
High-speed video frame S₁', S_Two', S_Three'.

FIG. 18 is a block diagram showing a high-speed video reproducing apparatus according to the fourth embodiment of the present invention. In the figure, the same reference numerals are given to the same parts as those in FIG. 1 (first embodiment), and the description is omitted. The high-speed video playback device according to the present embodiment further includes a cut point detection unit 501 corresponding to a break detection unit described in the claims, and the first high-speed video generation unit 2 included in the high-speed video playback device according to the first embodiment.
A fourth high-speed video generation unit 600 is provided instead of 00.

The cut point detecting section 501 detects a cut point in a video frame read from the video data storage section 101 via the video data reading section 103, and detects the cut point to indicate the cut point. The point position information CP is supplied to the fourth high-speed video generation unit 600. In the present embodiment, a point having a peak value is detected as a cut point based on the average luminance value of a video frame.

In the first to third embodiments, when generating a high-speed video frame for N × speed reproduction, N video frames are read from the video data storage unit 101. A total of 5N video frames (video frames FM _{+ 1-2N} to FM _{+ 3N} ) including the front and rear two frames in addition to the central N video frames are stored in the video data storage unit 101. Is read. If there are a plurality of cut points in 5N video frames, the target cut point is the cut point closest to the central N video frames.

When there is a cut point in the middle of a video frame, the fourth high-speed video generation section 600 performs averaging using only the video frame having the larger number of frames before and after the cut point, and outputs a high-speed video for high-speed reproduction. A frame is generated. The smaller number of video frames is used when generating high-speed video frames before and after the high-speed video frame generated from the larger number of video frames.

As shown in FIG. 19, the fourth high-speed video generation unit 600 is similar to the scene-linked frame selection unit 601 corresponding to the video frame selection means in the claims. 4 averaging unit 603. The scene-linked frame selection unit 601 uses the cut point position information CP sent from the cut point detection unit 501 to determine where and how many video frames are averaged.

For example, as shown in FIG. 20 (a), in addition to N central video frames, a total of 5N video frames including the front and rear two frames (video frames FM _{+ 1-2N)} To F _{M + 3N} ) are supplied to the fourth high-speed video generation unit 600, and when a high-speed video frame for 5 × -speed reproduction is generated around the video frame F _{M + (N + 1) / 2} , scene-linked The frame selection unit 601 selects a video frame used for averaging according to the position of the cut point.

First, as shown in FIG. 20 (b), when the cut point B1 is located after the video frame _{FM- (N-1) / 2} and before the video frame FM _{+ 1} , the cut point B1 The video frames from the subsequent video frame to the video frame FM _{+ N} are selected. Further, as shown in FIG.
There video frames from after the video frame _{F M + 1 F M + (} N + 1) / 2
, A video frame from the video frame after the cut point B2 to the video frame FM _{+ N} is selected.

As shown in FIG. 20D, when the cut point B3 is located after the video frame F _{M + (N + 1) / 2} and before the video frame F _{M + N} , the video frame F _{M +} The video frames from ₁ to the video frame before the cut point B3 are selected. Further, as shown in FIG.
Is the video frame FM _{+ (3N + 1)} after the video frame FM _{+ N.
) / 2,} the cut point B from the video frame FM _{+ 1}
The video frames up to the video frame before 4 are selected.

A scene-linked frame selection unit 601 will be described below.
Will be described in detail with reference to the flowchart in FIG. First, in step S701, the video frames FM _{+ 1-2N} to FM _{+ 3N are obtained} from the cut point position information CP sent from the cut point detecting unit 501.
Check if there is a cut point inside. Next, step S70
In 3, in accordance with the presence or absence of a cut point in step S701, if there is no cut point, the process proceeds to step S705, and if there is a cut point, the process proceeds to step S707.

In step S705, the video frame F
_{M + 1 to} FM _{+ N} are selected and the process is terminated. In step S707, if there is only one cut point, the position of the cut point is set as the cut point position BP, and if there are a plurality of cut points, the cut point closest to the video frame FM _{+ (N + 1) / 2} is set. The position is defined as a cut point position BP. Next, in step S709, it is checked whether or not the cut point position BP is between the video frame _{M- (N-1) / 2} and the video frame _{M + (N + 1) / 2.} If it is within the range, the process proceeds to step S711; otherwise, the process proceeds to step S713.

In step S711, the cut point position BP
After that, the video frames from the video frame F _{M + N} to the video frame F _{M + N} are selected, and the process ends. This flow is shown in FIG.
This corresponds to the examples shown in (b) and (c). In step S713, the cut point position BP is
_It is checked whether it is between _{M + (N + 1) / 2} and the video frame _{M + (3N + 1) / 2,} and if the cut point position BP is in this range, step S
The process proceeds to 715, and if not within this range, the process proceeds to step S717.

In the step S715, the video frame F
_{The process from M + 1} to the video frame before the cut point position BP is selected, and the process is terminated. This flow is shown in FIG.
This corresponds to the examples shown in (d) and (e). In step S717, the video frames F _{M + 1} to F
Select up to _{M + N} and end the process.

The fourth averaging section 603 averages the video frames selected by the scene-linked frame selecting section 601 in this way to generate a high-speed video frame. In the present embodiment, the fourth averaging unit 603 is
Although the description will be made assuming that the simple averaging described in the first embodiment is performed, the weighted averaging using the weighting coefficients described in the second embodiment may be performed. Also, as in the second example of the first embodiment, if the video data is subjected to the discrete cosine transform, the fourth averaging unit 603 sets the DCT
Averaging or weighted averaging may be performed on the DCT video frame represented by the coefficient.

As described above, in the high-speed video reproducing apparatus and the high-speed video reproducing method of the present embodiment, only high-speed video frames are generated by averaging video frames of the same scene. The video frame of the scene is not displayed. Therefore, it is possible to reproduce an easy-to-view high-speed video in which a scene can be clearly recognized. Therefore, even when there are many scenes, a user who visually searches for a high-speed video can easily perform the search.

In this embodiment, the cut point is detected in real time while the video frame is being read. However, information on the detection of the cut point is stored in advance, and the cut point is stored when high-speed playback is performed. Point position information CP
May be read out. At this time, the cut point detection unit 501 provided in the high-speed video playback device of the present embodiment is not necessary, and the read cut point position information CP is supplied to the scene-linked frame selection unit 601 of the fourth high-speed video generation unit 600. You.

In this embodiment, the cut point detecting section 5
Although 5N video frames are supplied to the 01 and fourth high-speed video generation units 600, 3N or 7N video frames may be supplied. In the first to fourth embodiments described above, the format of the video data is the RGB format, but may be the YUV format.

[0111]

As described above, according to the high-speed video reproducing apparatus, the high-speed video reproducing method and the recording medium of the present invention,
In a simple averaging means (simple averaging step), a plurality of continuous video frames constituting the video are averaged to generate a high-speed video frame for high-speed reproduction. Are displayed at regular intervals.

Therefore, even if the image is reproduced at a high speed, the information of all the video frames is reflected without any loss of information due to the thinning of the video frames, so that a video substantially similar to the original video can be reproduced smoothly and naturally at a high speed. In addition, even if the video contains fast moving objects, small objects, objects with a short appearance time, etc., the high speed video is reproduced by displaying the high speed video frame which averaged the video frame. The object is always reproduced. For this reason, a user who visually searches for a video can easily search for a desired scene at the same playback speed as compared with the related art, and can obtain a desired scene with the same accuracy and faster. , It is possible to search in a short time.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a high-speed video playback device according to a first embodiment of the present invention. It is.

FIG. 2 illustrates a first high-speed video generation unit according to the first embodiment.
FIG. 2 is a block diagram illustrating an internal configuration of the embodiment.

FIG. 3 is an explanatory diagram illustrating generation of a high-speed video frame by a first averaging unit according to the first embodiment.

FIG. 4 is an explanatory diagram illustrating continuous generation of high-speed video frames by a first averaging unit according to the first embodiment.

FIG. 5 is an explanatory diagram illustrating averaging of video frames performed by a first averaging unit according to the first embodiment.

FIG. 6 illustrates a second example of the first high-speed image generation unit according to the first embodiment.
FIG. 2 is a block diagram illustrating an internal configuration of the embodiment.

FIG. 7 is an explanatory diagram illustrating averaging of video frames performed by a second averaging unit according to the first embodiment.

FIG. 8 is an explanatory diagram illustrating an operation when averaging and reproducing a video frame in the high-speed video reproduction device according to the first embodiment.

FIG. 9 is a block diagram illustrating a high-speed video playback device according to a second embodiment of the present invention.

FIG. 10 is a block diagram illustrating an internal configuration of a second high-speed video generation unit according to the second embodiment.

FIG. 11 is an explanatory diagram illustrating generation of a high-speed video frame by a weighted averaging unit according to the second embodiment.

FIG. 12 is an explanatory diagram illustrating weighted averaging of a video frame performed by a weighted averaging unit according to the second embodiment.

FIG. 13 is an explanatory diagram showing an original video frame, a high-speed video frame generated by averaging in the first embodiment, and a high-speed video frame generated by weighted averaging in the second embodiment.

FIG. 14 is a block diagram illustrating a high-speed video playback device according to a third embodiment of the present invention.

FIG. 15 is a block diagram showing an internal configuration of a third high-speed video generator of the third embodiment.

FIG. 16 is an explanatory diagram showing a plurality of video frames and a high-speed video frame generated according to the third embodiment.

FIG. 17 is an explanatory diagram showing a plurality of video frames, a high-speed video frame generated according to the first embodiment, and a high-speed video frame generated according to the third embodiment.

FIG. 18 is a block diagram illustrating a high-speed video playback device according to a fourth embodiment of the present invention.

FIG. 19 is a block diagram showing the internal configuration of a fourth high-speed video generation unit according to the fourth embodiment.

FIG. 20 is an explanatory diagram illustrating 5N video frames and a rule of selecting a video frame to be averaged according to a cut point.

FIG. 21 is a flowchart illustrating a method of selecting a video frame performed by a scene-linked frame selection unit.

FIG. 22 is an explanatory diagram illustrating a fast-forward playback method of a VTR.

FIG. 23 is an explanatory diagram showing an image displayed by the VTR fast-forward playback method.

FIG. 24 is an explanatory diagram for explaining a frame skipping reproduction method of a digital video.

[Explanation of symbols]

 Reference Signs List 101 Video data storage unit 103 Video data reading unit 105 Video reproduction unit 200 First high-speed video generation unit 201 First averaging unit 203 Variable length decoding (VLD) unit 205 Inverse quantization unit 207 Second averaging unit 209 Inverse DCT unit 300 Second high-speed video generation unit 301 Weighted averaging unit 303 Multiplication unit 305 Addition unit 307 Division unit 400 Third high-speed video generation unit 401 Video frame division unit 403 Region change detection unit 405 Third averaging Unit 407 video data synthesis unit 501 cut point detection unit 600 fourth high-speed video generation unit 601 scene-linked frame selection unit 603 fourth averaging unit

Claims (17)

[Claims] 1. A simple averaging means for averaging a plurality of continuous video frames constituting a video to generate a high-speed video frame for high-speed reproduction, and a constant high-speed video frame generated by the simple averaging means. A high-speed video playback device comprising: video playback means for displaying at intervals. 2. A weighted averaging means for weighing each of a plurality of continuous video frames forming a video, averaging the plurality of video frames, and generating a high-speed video frame for high-speed reproduction, A high-speed video playback device, comprising: video playback means for displaying the high-speed video frames generated by the averaging means at regular intervals. 3. A video frame dividing unit which divides a plurality of continuous video frames constituting a video into a plurality of regions, respectively; Region change detection means for detecting a degree of change; and weighted averaging for averaging or weighting the plurality of video frames for an area in which the degree of change detected by the area change detection means is equal to or greater than a predetermined level. Averaging means for performing the averaging or weighted averaging by the averaging means, and video of a predetermined video frame in an area where the degree of change detected by the area change detecting means is smaller than a predetermined level. Video data synthesizing means for synthesizing data and generating a high-speed video frame for high-speed reproduction; Fast video reproducing apparatus characterized by comprising: a video reproduction means, the displaying fast video frame at regular intervals. 4. The high-speed video reproducing apparatus according to claim 3, wherein said area change detecting means detects a degree of change based on an average of hue or luminance of a video frame. 5. The high-speed image represented by a transform coefficient, wherein the simple averaging means, the weighted averaging means or the averaging means averages or weights a plurality of orthogonally coded video frames. 5. The high-speed video reproducing apparatus according to claim 1, further comprising: an inverse orthogonal transform unit for generating a frame and performing an inverse orthogonal transform on the high-speed video frame represented by the transform coefficient. 6. When reproducing a video at N times speed, the simple averaging means, the weighted averaging means or the averaging means averages or weights the N video frames to obtain one high speed video frame. 6. The high-speed video playback device according to claim 1, wherein the video frame is generated. 7. When there is a scene break in the middle of a predetermined number of continuous video frames constituting a video, the simple averaging means, the weighted averaging means or the averaging means averages according to a break position. 6. The high-speed video reproducing apparatus according to claim 1, further comprising a video frame selecting means for selecting a video frame to be weighted and averaged. 8. The high-speed video reproducing apparatus according to claim 7, further comprising a delimiter detecting means for detecting a delimiter of a scene in the predetermined number of video frames. 9. A simple averaging step of averaging a plurality of continuous video frames constituting a video to generate a high-speed video frame for high-speed reproduction, and fixing the high-speed video frame generated in the simple averaging step to a constant value. A high-speed video playback method, comprising: a video playback step of displaying at intervals. 10. A weighted averaging step of assigning weights to each of a plurality of continuous video frames constituting a video, averaging the plurality of video frames, and generating a high-speed video frame for high-speed playback; A video playback step of displaying the high-speed video frames generated in the averaging step at regular intervals. 11. A video frame dividing step of dividing each of a plurality of continuous video frames constituting a video into a plurality of regions, and for each of the regions divided in the video frame dividing step, the plurality of video frames extend over the plurality of regions. An area change detection step of detecting a degree of change, and averaging or weighting the plurality of video frames with respect to an area in which the degree of change detected in the area change detection step is equal to or greater than a predetermined level. Averaging or weighted averaging in the averaging step, and video of a predetermined video frame in an area where the degree of change detected in the area change detection step is smaller than a predetermined level. Synthesizing the video data to generate a high-speed video frame for high-speed playback; Fast video reproducing method characterized in that it comprises a video reproducing step of displaying a high-speed video frame generated by the synthesis step at regular intervals, the. 12. The high-speed video reproduction method according to claim 11, wherein the area change detection step detects a degree of change based on an average of hue or luminance of the video frame. 13. The high-speed image represented by a transform coefficient, wherein the simple averaging step, the weighted averaging step, or the averaging step averages or weights a plurality of orthogonally coded video frames. 13. The high-speed video reproduction method according to claim 9, further comprising an inverse orthogonal transforming step of generating a frame and performing an inverse orthogonal transform on the high-speed video frame represented by the transform coefficient. 14. When reproducing an image at N times speed at a high speed, the simple averaging step, the weighted averaging step or the averaging step averages or weights and averages N image frames to obtain one high-speed image. The video frame is generated, wherein the video frame is generated.
3. The high-speed video reproduction method according to 3. 15. When there is a scene break in the middle of a predetermined number of continuous video frames constituting a video, the simple averaging step, the weighted averaging step or the averaging step averages according to a break position. Or a video frame selecting step of selecting a video frame to be weighted and averaged.
The high-speed video reproduction method according to 0, 11, 12 or 13. 16. The high-speed video playback method according to claim 15, further comprising a break detection step of detecting a break of a scene in the predetermined number of video frames. 17. The method of claim 9, 10, 11, 12, 13,
A computer-readable recording medium that records a program for causing a computer to execute the high-speed video reproduction method according to 14, 15, or 16.