JP2825079B2 - Moving image generation method and encoding / decoding method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a moving picture generation method and an encoding / decoding method.

[0002]

2. Description of the Related Art Different from a waveform encoding apparatus, there have been proposed some schemes for reproducing an approximate moving image by using a characteristic amount unique to an image signal. As an example of a conventional technique for generating a moving image from only simple motion parameters (motion information),
The present inventor has already disclosed in Japanese Unexamined Patent Application Publication No. 7-203409 (Japanese Patent Application No. 5-33391) that an image sequence is output according to an extracted motion parameter to generate a reproduced moving image, so that only a simple motion parameter is used. As a method of reproducing a moving image, parameter extracting means for extracting a main motion parameter from motion information of the moving image, image sequence storing means for storing an image sequence corresponding to the motion parameter, There is proposed a moving image reproducing apparatus including: a generating unit that outputs an image sequence from an image sequence storing unit to generate a reproduced moving image. That is, the publication describes a method of analyzing the motion of a subject, using a motion parameter representing the extracted motion as a code, and reproducing the same motion as the subject based on the code.

[0003] In this case, the motion of the subject is extracted as follows.
In addition to a method of extracting a subject portion from an image and obtaining a motion vector thereof, a method of analyzing information representing motion included in audio information or text, or the like can be used.

[0004] On the other hand, the reproduction of motion is performed by smoothly connecting video scenes stored in advance. As the stored data, typical scenes are recorded for each application. For example, if it is limited to the use of a videophone, the upper body image of a person is a scene such as greeting, talking, and nodding.

[0005]

By the way, as long as the decoder has a large video memory and can store all of various scenes, any motion can be reproduced. In many cases, only a very limited number of scenes can be stored in the video memory due to restrictions or the like.

[0006] Since the moving image reproduced by the conventional method is limited to the stored video scene,
Due to cost constraints, only limited movements can be reproduced.

In applications such as communication, real-time processing is required. However, it is difficult to automatically analyze complicated movements such as smiles in real time, and in practice, extremely simple operations are performed. At present, it can only be reproduced.

Accordingly, the present invention has been made in view of the above circumstances, and an object of the present invention is to express a complicated motion by a plurality of simple motion parameter strings to thereby store a limited amount of stored moving images. An object of the present invention is to provide a method for reproducing various moving images from images. Further, an object of the present invention is to automatically generate a motion parameter expressing a complex motion,
An object of the present invention is to provide a method for reproducing various operations that have been difficult to analyze in image processing.

[0009]

In order to achieve the above object, the present invention provides a method of generating a motion parameter sequence from an index indicating a predetermined operation and current position information,
Means for generating position information at a fixed time interval in a predetermined motion trajectory represented by the index, means for exclusively selecting one of the generated position information and the current position information, Means for generating a motion parameter sequence from the obtained position information.

The outline of the present invention will be described below. For example, in the case of a motion of a person, a portion common to a plurality of different motions often exists. Therefore, it is possible to generate a complicated movement by connecting common parts that are simple movements. As a result, these operations can be reproduced without accumulating all the various operations independently. Also, instead of recognizing complicated operations by complicated image processing, a button for identifying image information (as a simple example, an event such as pressing an alphanumeric button or clicking an icon is used as an index of a specific subject operation). ) Or a sensor output that measures the movement of a subject and passed to the system as an index signal, the system can be constructed with an extremely simple configuration. For example, when the motion of a subject is measured by a sensor, a motion index is determined based on the direction and amount of the motion calculated from the sensor output, and a specific motion motion is determined based on natural language analysis or the like from text information or voice information. May be analyzed, and this may be used as an index to indicate the operation content.

As described above, in the first motion parameter sequence generation method according to the first aspect of the present invention, there is provided a means for generating position information of a trajectory of a complicated motion, and a method for generating position information of the generated position information and current position information. It comprises means for selecting one of them, and means for generating a motion parameter from the selected position information.

According to a second motion parameter sequence generation method according to the second aspect of the present invention, in addition to the first motion parameter generation method, a position information of a trajectory is used for smoothing a transition from a current position to a trajectory position information. It further includes means for adding the location information.

According to a third motion parameter generation method according to the third aspect of the present invention, in addition to the first motion parameter generation method, the position information of the trajectory smoothly transitions to the current position after the generation of the motion parameter. And means for adding post-position information for performing the operation.

According to a fourth aspect of the present invention, in addition to the first motion parameter generation method, the fourth motion parameter sequence generation method is used to smooth transition from the current position to the position information of the trajectory. And means for adding post-position information for smoothly transitioning to the current position after the generation of the motion parameter.

According to a fifth aspect of the present invention, there is provided a moving image reproduction system, comprising: a means for calculating current position information from a motion parameter in addition to any one of the first to fourth motion parameter sequence generation systems; Means for reproducing a moving image from the motion parameters.

According to a sixth aspect of the present invention, in the moving picture encoding / decoding method, the encoder extracts a motion parameter from an original signal including motion information, and the first to fourth ones. The decoder includes a motion parameter sequence generating unit and a unit that tracks the extracted motion to determine the current position, and the decoder includes a unit that reproduces a moving image from the motion parameters.

According to a seventh aspect of the present invention, there is provided the moving picture encoding / decoding method, wherein the encoder extracts a motion parameter from an original signal including motion information, and executes the extracted motion parameter and a specific operation. The decoder comprises means for multiplexing the sign indicating the meaning into a code, and the decoder separates the motion parameter extracted from the multiplexed code into the index indicating a specific operation; 4; a means for calculating current position information from the separated motion parameters; and a means for reproducing a moving image from the motion parameters.

[0018]

Embodiments of the present invention will be described below with reference to the drawings.

[0019]

Embodiment 1 An embodiment of a first motion parameter generation method according to the invention described in claim 1 (“First Embodiment”)
) Will be described with reference to FIG. FIG. 1 illustrates this first
FIG. 3 is a diagram showing a configuration of a motion parameter sequence generation method of FIG.

Referring to FIG. 1, a motion parameter generation circuit 10 includes a motion index 101 given as an input.
A motion parameter sequence 105 representing a series of operations indicated by an index from the current position information 103 and the current position information 103. The characteristic point position information generation circuit 11, the position information changeover switch 13, and the operation parameter generation circuit It consists of 12.

The operation of this embodiment will be described below.

The motion index 101 is an index indicating various complicated motions, and is converted by the characteristic point position information generation circuit 11 into characteristic point position information 102 which is a locus of the motion at regular time intervals.

The feature point position information 102 is composed of a vector P (i) (i = 1, 2,..., N), which is a sequence of N vector data representing positions (sampling time points). Here, N is a parameter indicating the length of the operation, and is uniquely determined by the operation index 101.

The generation of the position information 102 of the feature point is performed by storing the vector P (i) (i = 1, 2,..., N) in a memory (not shown) and reading the vector P (i) from an index signal. , Only the position information of the feature points at indefinite time intervals is stored, and the moving speed is used as a parameter,
A method of generating by interpolation using a method such as linear or spline interpolation, or a method of storing only coarsely sampled position data and generating position data at regular time intervals by a similar interpolation method can be used. .

The generated feature point position information 102 is input to the position information changeover switch 13 and the feature point position information 1
02, the feature point position information 102 is displayed during the period in which the current
Is output as That is, while the feature point position information 102 is being output from the feature point position information generation circuit 11, the position information changeover switch 13 is connected to the terminal “a” and the feature point position information 102 is transmitted to the operation parameter generation circuit 1.
When the feature point position information 102 is not output, the terminal is switched to the terminal b side, and the current position information 103 is output to the operation parameter generation circuit 12.

Here, the current position information 103 is generated at regular time intervals by vector data representing the current position information. Therefore, a part (N pieces) of the current position information 103 is replaced with the feature point position information 102.

The selected position information 104 is input to the motion parameter generation circuit 12 and calculates the difference between successive vectors to generate the motion parameter 1 generated.
05 is calculated. When obtaining the difference between successive vectors, for example, if the movement amount (difference amount) is too large, motion control such as suppressing the difference amount is performed.

The generated motion parameters 105 are vector data generated at regular time intervals.

The operation index 101 may be generated by natural language analysis from text information or the like, or may be generated by pressing a button or the like. For example, based on the natural language "laughing", this may be used as a motion index of a moving image to laugh, or an index corresponding to the button may be generated by pressing a button corresponding to the moving image. Is also good.

Since one motion index generates one or more motion parameters, the next motion index may be input before the previous motion index is completed in some cases. At this time, the generation of the motion parameter corresponding to the previous index is immediately terminated when a new motion index is input, and the processing of the next index is performed.

[0031]

[Embodiment 2] An embodiment of a second motion parameter generation method (referred to as "second embodiment") according to the second aspect of the present invention will be described below with reference to FIG. FIG. 2 is a diagram showing a configuration of the second motion parameter generation method. Referring to FIG. 2, the present embodiment has a configuration in which a prefix position information insertion circuit 14 is added to the first motion parameter generation method shown in FIG.

The operation of this embodiment will be described with reference to FIG.

The motion parameter generation circuit 10 of the present embodiment
In addition to the operation similar to that of the first embodiment, the transition from the current position can be performed smoothly before the characteristic point position information 102 at fixed time intervals generated by the characteristic point position information generation circuit. First, the front position information is inserted using the front position information insertion circuit 14. That is, in the present embodiment, instead of directly inputting the feature point position information 102 to the position information changeover switch 13, the front position information obtained by modifying the feature point position information 102 by the front position information insertion circuit 14 is used. This is for inputting the feature point position information 106 including the information. Otherwise, the operation is the same as in the first embodiment. The position information changeover switch 13 is connected to the terminal a while the feature point position information 106 is being output from the preceding position information insertion circuit 14, and outputs the feature point position information 106 to the operation parameter generation circuit 12, When the feature point position information 106 is not output from the prefix position information insertion circuit 14, the terminal is switched to the terminal b side, and the current position information 103 is output to the operation parameter generation circuit 12.

The position information 104 selected by the position information changeover switch 13 is input to a front position information insertion circuit 14, and the front position information insertion circuit 14
2 and the selected position information 104, feature point position information 106 including prefix position information is generated.

Vector P (i) of feature point position information 102
(I = 1, 2,..., N) are generated only from the motion index 101 and replace a part of the current position information 103. In the first embodiment, therefore, the feature point position information is obtained from the current position information 103. Continuity at the moment of switching to 102 is not guaranteed.

Therefore, in the present embodiment, when the feature point position information 102 is input, the vector output to the selected position information 104 is checked, and the previously output selected position information 104 is replaced with the initial position information. As an information vector P _INIT , a vector P _p (i) (i = 1, 2,...) Of front position information for interpolating between the vector P _INIT and the vector P (1)
N _p ) is generated and is added before the feature point position information 102.

That is, the ｛vector P _p (i) (i = 1,
2,..., N _p ), vector P (i) (i = 1, 2,.
N) The sum of N _p + N vectors of 特 徴 is output as feature point position information 106 including prefix position information.

As a result, in this embodiment, N _p
× (time interval of vector data) delay occurs,
Motion parameters that are smoother than in the first embodiment can be generated.

[0039]

Third Embodiment A third motion parameter generation method (hereinafter, referred to as a "third embodiment") according to an embodiment of the present invention will be described with reference to FIG. FIG. 3 is a diagram showing a configuration of the third motion parameter generation method. This embodiment is different from the configuration of the first embodiment shown in FIG.
A post-position information insertion circuit 15 is added. That is, in the present embodiment, the operation index 101
, A position information generation circuit 11, a position information switch 13, and a position information generation circuit 1 for a characteristic point.
, A post-position information insertion circuit 15 which receives the output 102 of the No. 1 and the current position information 103 as inputs, and a post-position information insertion circuit 15
A position information switch 13 for switching the output 107 of the current position information and the current position information 103 and an operation parameter generation circuit 12 are provided.

Next, the operation of this embodiment will be described with reference to FIG.

The motion parameter generation circuit according to the present embodiment
In addition to the same operation as in the first embodiment, after the generated feature point position information 102 at a fixed time interval, the post-position information is post-fixed so that the transition to the current position can be performed smoothly after the parameter generation. It is inserted using the position information insertion circuit 15. The position information changeover switch 13 is connected to the terminal a while the position information 107 is being output from the post-position information insertion circuit 15, outputs the position information 107 to the operation parameter generation circuit 12, and outputs the position information 107. When the position information 107 is not output from the insertion circuit 15, the terminal b is switched to the terminal b side, and the current position information 103 is changed to the operation parameter generation circuit 1.
Output to 2.

The operation of this embodiment is based on the feature point position information 10
2 is input directly to the position information changeover switch 13, but instead of the feature point position information 10
Feature point position information 1 including postposition position information with reorganization added to 2
The operation is basically the same as that of the first embodiment except for the point that 07 is input.

The operation of the post-position information insertion circuit 15 will be described below. The post-position information insertion circuit 15 generates feature point position information 107 including post-position information from the feature point position information 102 and the current position information 103.

As described above, the vector P (i) (i = 1, 2,..., N) of the feature point position information 102 is generated only from the motion index 101 and replaces a part of the current position information 103. In the first embodiment, continuity at the moment of switching from the feature point position information 102 to the current position information 103 is not guaranteed.

Therefore, in this embodiment, after outputting all the feature point position information 102 to the position information changeover switch as the feature point position information 107 including the post-position information, the vector P _TERM of the current position information 103 is checked, and Point location information 1
It is checked whether the difference between the last vector P (N) of 02 and the vector _PTERM is smaller than a predetermined threshold.

[0046] If the difference between the vector P (N) and the vector P _TERM is smaller than the threshold, it will not add any effect, if large, connecting vector P _{TE RM} and vector P (N) The feature point position information 107 including the postposition position information which has transmitted the vector P _TEMP existing on the line segment so far.
Output after.

The operation of adding the new vector is continued until the difference between the vector of the feature point position information 107 including the postposition information output last and the vector _PTERM becomes smaller than a predetermined threshold value.

As a result, the return to the current position information
Although a delay of (the number of additions) × (time interval of vector data) occurs, it is possible to generate smoother motion parameters than in the first embodiment.

[0049]

Fourth Embodiment A fourth motion parameter generation method (hereinafter referred to as a "fourth embodiment") according to an embodiment of the present invention will be described with reference to FIG. FIG. 4 is a diagram showing a configuration of the fourth motion parameter generation method. The motion parameter generation circuit 10 of the present embodiment is different from the configuration of the first embodiment shown in FIG. 1 in that the front position information insertion circuit 14 and the post position information described in the second and third embodiments are used. The configuration is such that an insertion circuit 15 is added. That is, in the fourth motion parameter sequence generation method, the feature point position information generation circuit 11 which receives the motion index 101 as an input is used.
A position information changeover switch 13, a feature point position information generation circuit 11, and an output 10 of the position information changeover switch 13.
4 as an input, a post-position information insertion circuit 15 receiving the output of the pre-position information insertion circuit 14 and the current position information 103, and a post-position information insertion circuit 15 A position information switch 13 for switching between the output 108 of the current position information and the current position information 103, and an operation parameter generation circuit 12.

The operation of the present embodiment will be described below with reference to FIG.

In the present embodiment, in addition to the same operation as that of the first embodiment, a prefix is provided before the generated characteristic point position information 102 at a fixed time interval so that the transition from the current position can be performed smoothly. The post-position information is inserted by the post-position information insertion circuit 15 so that the position information is inserted using the front-position information insertion circuit 14 and, after the parameters are generated, the position information can be smoothly shifted to the current position.

The operation of this embodiment is based on the feature point position information 10
2 is input directly to the position information changeover switch 13, and feature point position information including rear position information obtained by modifying the characteristic point position information 102 by the front position information insertion circuit 14 and the rear position information insertion circuit 15. The other points are basically the same as those of the first embodiment except that the point of inputting is 108. The position information changeover switch 13 is connected to the terminal a while the characteristic point position information 108 is being output from the post-position information insertion circuit 15 and is connected to the terminal a.
Is output to the operation parameter generation circuit 12, and when the feature point position information 108 is not output from the post-position information insertion circuit 15, the terminal b is switched to the terminal b side, and the current position information 103 is output to the operation parameter generation circuit 12.

The operation of the prefix position information insertion circuit 14 in the present embodiment is the same as that of the second embodiment, but in the second embodiment (see FIG. 2), the output 106 is used for switching the position information. What was input to the switch 13
The input has been changed to the post-position information insertion circuit 15.

The operation of the post-position information insertion circuit 15 in the present embodiment is the same as that of the third embodiment, but in the third embodiment (see FIG. 3), The position information 10 of the feature point input to the circuit 15
2 has been changed to the feature point position information 106 including the prefix position information output from the prefix position information insertion circuit 14.

As a result, the transition from the current position information and the return to the current position after the parameter is generated become smoother than in the first embodiment.

[0056]

[Embodiment 5] FIG. 5 is a diagram showing a configuration of an embodiment of the invention according to claim 5, and shows an embodiment of an apparatus for realizing an image reproduction method (referred to as "fifth embodiment"). FIG. With reference to FIG. 5, this apparatus obtains a reproduced moving image 201 from a motion index 101 given as an input.
And includes a motion parameter generation circuit 10, a position determination circuit 21 for calculating the current position, and a moving image reproduction circuit 22. Motion parameter generation circuit 10
The motion parameter generation circuit 10 according to any one of the first to fourth embodiments described with reference to FIGS. 1 to 4 is used.

The operation of the present embodiment will be described below with reference to FIG.

Operation index 101 and current position information 2
Using 02, the motion parameter generation circuit 10 generates the motion parameter sequence 105. The generated motion parameter sequence 105 is sent to the moving image reproducing circuit 22 to generate a reproduced moving image 201. At the same time, the generated motion parameter sequence 105 is sent to the position determination circuit 21 and is used for calculating the current position 202 by integrating the past generated motion parameter sequence 105.

The moving image reproducing circuit 22 is disclosed in, for example, Japanese Patent Application Laid-Open No. 7-203409 (Japanese Patent Application No. 5-33391).
Can be used. That is, as the moving image reproduction circuit 22, for example, an image sequence storage circuit (not shown) storing image sequences having various motions,
An image sequence necessary for reproducing a moving image in accordance with the motion parameter sequence from the motion parameter generating circuit 10 is received from an image sequence storage circuit, and a moving image generating circuit (not shown) that reproduces the moving image using the image sequence is used. You. In this case, it is assumed that the image sequence storage circuit previously stores motion images in each direction necessary for reproducing the motion image from the input motion information, and is limited to, for example, videophone use. Then, in the upper body image of the person, an image such as facing right, facing left, or speaking is stored. Also, a moving image reproduction circuit 2
The second moving image generation circuit (not shown) generates the reproduced moving image 201 based on the motion parameters input from the motion parameter generation circuit 10. For example, the motion of the entire object is given as the motion parameter 105. In this case, the motion is performed by generating a motion sequence from the image one frame before to a place where the image should move according to the motion parameter 105, and synthesizing the motion sequence with the previously reproduced moving image. That is, to generate a motion sequence, a control signal indicating which sequence is necessary is sent to an image sequence storage circuit storing various motion sequences, and the image sequence storage circuit The sequence in the storage device is searched based on the signal, and a predetermined motion sequence is sent back to the moving image generation circuit.

[0060]

[Embodiment 6] FIG. 6 is a diagram showing a configuration of an embodiment of the invention according to claim 6, showing an embodiment of a moving picture coding and decoding apparatus (referred to as "sixth embodiment"). FIG. With reference to FIG. 6, in the present embodiment, the encoder 51 includes:
It outputs a sequence of motion parameters 105 generated from an original signal 301 including a motion index 101 and motion information given as inputs, and includes a motion parameter extraction circuit 31, a motion parameter generation circuit 10, a position determination circuit 21, Consists of The decoder 52 outputs a reproduced moving image 201 from the generated motion parameter sequence 105 given as an input.
Composed of

The operation of the present embodiment will be described below with reference to FIG.

The encoder 51 is preferably disclosed in Japanese Patent Application Laid-Open No. Hei 7-203409 (Japanese Patent Application No. 5-33391).
And a code that is compatible with the encoder of the moving image encoding / decoding device described in (1). Note that the encoder described in the publication includes a motion parameter extraction circuit that extracts main parameters from motion information, and the encoding / decoding device described in the publication uses an existing inter-frame prediction encoding device (for example, an ITU-T It reproduces a moving image that is compatible with Recommendation H.261 etc.). In the present embodiment, as described above, preferably, a code that can reproduce even a complicated operation pattern specified by a button or the like may be generated.

In the encoder 51, the original signal 301 including the motion information is converted into a motion parameter extracting circuit 31 for extracting main parameters from the motion information (for details, see Japanese Patent Laid-Open No. 7-2034).
No. 09 gazette) and the motion parameters 30
2 is extracted.

The motion parameters 302 extracted by the motion parameter extraction circuit 31 are input to the position determination circuit 21 and current position information 303 is generated by integrating the past motion parameters 302.

Current position information 303 and operation index 1
01 is input to the motion parameter generation circuit 10 configured in any of the first to fourth embodiments, and a motion parameter sequence 105 is generated.

In the decoder 52, the moving image reproducing circuit 22 converts the received moving parameter sequence 105 into the reproduced moving image 2.
01 is output. That is, the decoder 52 stores an image sequence required to reproduce a moving image in accordance with a motion parameter sent from the encoder, and an image sequence storage circuit storing an image sequence having various motions. A moving image reproducing circuit including a moving image generating circuit that receives a signal from the circuit and reproduces a moving image by using the circuit (for details, refer to Japanese Patent Application Laid-Open No. 7-203409).

[0067]

[Embodiment 7] FIG. 7 is a diagram showing a configuration of an embodiment of the present invention according to claim 7, showing an embodiment of a moving picture coding and decoding apparatus (referred to as "seventh embodiment"). FIG. Referring to FIG. 7, in the present embodiment, encoder 5
3 outputs multiplexed motion information 401 from the motion index 101 given as an input and the original signal 301 containing motion information, and comprises a motion parameter extraction circuit 31 and a multiplexing circuit 41. .

The decoder 54 outputs a reproduced moving image 201 from the multiplexed motion information 401 given as an input.
And comprises a demultiplexing circuit 42, a motion parameter generation circuit 10, a position determination circuit 21, and a moving image reproduction circuit 22.

The operation of the present embodiment will be described below with reference to FIG.

The encoding and decoding unit generates a reconstructed moving image.
Although it is not compatible with the method described in JP-A-09-0909, it generates a more compact code.

In the encoder 53, the original signal 301 including the motion information is analyzed by the motion parameter extracting circuit 31, and the motion parameter 302 is extracted, as described in Japanese Patent Laid-Open No. 7-203409.

The extracted motion parameters 302 are multiplexed by the multiplexing circuit 41 with the motion index 101 given as an input, and multiplexed motion information 401 is output.

In the sixth embodiment, the encoder 5
While the motion parameter sequence 105 output from the motion parameter generation circuit 10 is transmitted to the decoder 52, in the present embodiment, the first
The motion parameter generation circuit 10 described in the fourth to fourth embodiments is provided, and the motion index developed into a plurality of motion parameters from the encoder 63 is directly incorporated into the code. Thus, a more compact code can be generated.

In the decoder 54, the demultiplexing circuit 42
However, the multiplexed motion information 401 is separated into the separated motion index 402 and the separated motion into parameters 403.

The separated motion parameters 403 are input to the position determination circuit 21, and the current position information 303 is generated by integrating the past motion parameters 403.
The generated current position information 303 and the separated motion index 402 are input to the motion parameter generation circuit 10 having any one of the first to fourth embodiments, and the motion parameter sequence 105 is generated and output. You. The motion parameter sequence 105 is sent to the moving image reproducing circuit 22 to generate a reproduced moving image 201. Moving image reproduction circuit 2
2 receives an image sequence storage circuit (not shown) storing an image sequence having various motions, and an image sequence necessary for reproducing a moving image from the image sequence storage circuit according to the motion parameters, and uses the received image sequence. And a moving image generation circuit (not shown) for reproducing the reproduced moving image 201.

[0076]

As described above, according to the first aspect of the present invention, a simple operation necessary for reproducing a complicated operation based on a button input or the like is performed without using an expensive sensor or image processing. It is possible to generate a simple motion parameter sequence. Further, according to the present invention, there is an effect that accumulated data required for reproducing a plurality of operations having a common portion can be reduced.

According to the second to fourth aspects of the present invention, similar to the first aspect of the present invention, a simple motion parameter sequence required to reproduce a complicated operation is formed based on a button input or the like. Not only reduce the amount of stored data required to reproduce multiple operations that have common parts, but also obtain a smooth image when switching the generated data. Have.

Further, according to the moving image reproducing method according to the fifth aspect of the present invention, there is an effect that a moving image which performs a complicated movement specified by a button input or the like can be created.

Further, according to the moving picture coding / decoding system according to the invention described in claim 6, the above-mentioned Japanese Patent Laid-Open No. 7-2034 is disclosed.
It is possible to generate a code that can be decoded by using the moving picture decoder described in Japanese Patent Application Laid-Open No. 09-0909, and has an effect that it is possible to reproduce a complicated operation that has conventionally been difficult to detect.

According to the moving picture coding / decoding system according to the invention described in claim 7, a more compact moving picture coding / decoding system disclosed in the above-mentioned Japanese Patent Application Laid-Open No. 7-203409 is used. The code has the advantage that complex operations can be reproduced.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration of a motion parameter generation method according to a first embodiment of the present invention.

FIG. 2 is a diagram illustrating a configuration of a motion parameter generation method according to a second embodiment of the present invention.

FIG. 3 is a diagram illustrating a configuration of a motion parameter generation method according to a third embodiment of the present invention.

FIG. 4 is a diagram illustrating a configuration of a motion parameter generation method according to a fourth embodiment of the present invention.

FIG. 5 is a diagram showing a configuration of a moving image reproduction method according to a fifth embodiment of the present invention.

FIG. 6 is a diagram illustrating a configuration of an image encoding / decoding method according to a sixth embodiment of the present invention.

FIG. 7 is a diagram illustrating a configuration of an image encoding / decoding method according to a seventh embodiment of the present invention.

[Explanation of symbols]

 Reference Signs List 10 motion parameter generation circuit 11 feature point position information generation circuit 12 operation parameter generation circuit 13 position information switching circuit 14 front position information insertion circuit 15 post position information insertion circuit 21 position determination circuit 22 moving image reproduction circuit 31 motion parameter extraction Circuit 41 Multiplexing circuit 42 Demultiplexing circuit 101 Motion index 102 Feature point position information 103 Current position information 104 Selected position information 105 Generated motion parameter sequence 106 Feature point position information including prefix information 107 Postfix information Included feature point position information 108 Feature point position information including preceding and following information 201 Reproduced moving image 202 Current position information 301 Original signal including motion information 302 Extracted motion parameters 303 Current position information 401 Multiplexed motion information 402 Isolated motion index 403 Separated extracted motion information

Claims (7)

(57) [Claims] 1. A method of generating a motion parameter sequence from an index indicating a predetermined operation and current position information, wherein the unit generates position information at regular time intervals on a trajectory of the predetermined operation represented by the index. And means for exclusively selecting any of the generated position information and the current position information; and means for generating a motion parameter sequence from the selected position information. Motion parameter sequence generation method. 2. A method for generating a motion parameter sequence from an index indicating a predetermined operation and current position information, comprising: means for generating position information at regular time intervals in a trajectory of a predetermined operation represented by the index. Means for adding prefix position information to the generated position information using the selected position information; position information at a fixed time interval to which the prefix position information is added; and the current position information. A means for exclusively selecting any one of the following: and a means for generating a motion parameter sequence from the selected position information. 3. A method for generating a motion parameter sequence from an index indicating a predetermined operation and current position information, wherein the unit generates position information at regular time intervals in a trajectory of the predetermined operation represented by the index. Means for adding post-position information to the generated position information using the current position information; position information at a fixed time interval to which the post-position information is added; and the current position information A means for exclusively selecting any one of the following: and a means for generating a motion parameter sequence from the selected position information. 4. A method for generating a motion parameter sequence from an index indicating a predetermined operation and current position information, wherein the position information is generated at regular time intervals in a trajectory of the predetermined operation represented by the index. Means for adding prefix position information to the generated position information using the selected position information; and position information at a fixed time interval to which the prefix position information is added, the current position information Means for adding post-position information using, the position information at a fixed time interval to which the pre-position information and the post-position information are added, or any of the current position information exclusively A method for generating a motion parameter sequence, comprising: means for selecting; and means for generating a motion parameter sequence from the selected position information. 5. A method for reproducing a moving image from an index instructing a predetermined operation, comprising: generating a motion parameter sequence expressing an operation specified by the index from the index and current position information; 5. The motion parameter sequence generation method according to claim 1, a unit that calculates current position information from the generated motion parameters, and a unit that reproduces a moving image from the generated motion parameters. 6. A moving picture reproduction method, comprising: 6. An encoder for encoding a moving image from an original signal including motion information and an index for instructing a predetermined operation, and a decoder for reproducing the moving image with an output from the encoder as an input. An encoding and decoding method for a moving image, comprising: a unit for extracting a motion parameter from the original signal; and a unit for calculating current position information from the extracted motion parameter. 5. The motion parameter sequence generation method according to claim 1, wherein a motion parameter sequence is generated from the index and the calculated current position information. 6. A moving image encoding / decoding method, comprising: means for reproducing a moving image from a set motion parameter. 7. An encoder for encoding a moving image from an original signal including motion information and an index for instructing a predetermined operation, and a decoder for reproducing the moving image by using an output from the encoder as an input. An encoding and decoding method for a moving image, comprising: means for extracting a motion parameter from the original signal; and multiplexing the extracted motion parameter and the index to generate a code. Means for separating a motion parameter and an index indicating a predetermined operation from the multiplexed code, and calculating current position information from the separated motion parameter. Means for generating a motion parameter sequence from the separated indices and the calculated current position information, and means for reproducing a moving image from the generated motion parameters. Encoding and decoding system of the video to symptoms.