JPH05336510A - Motion compensation encoding device and decoding device - Google Patents

Motion compensation encoding device and decoding device

Info

Publication number
JPH05336510A
JPH05336510A JP14246592A JP14246592A JPH05336510A JP H05336510 A JPH05336510 A JP H05336510A JP 14246592 A JP14246592 A JP 14246592A JP 14246592 A JP14246592 A JP 14246592A JP H05336510 A JPH05336510 A JP H05336510A
Authority
JP
Japan
Prior art keywords
motion
motion vector
signal
decoding
motion compensation
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
JP14246592A
Other languages
Japanese (ja)
Inventor
Shinya Sumino
眞也 角野
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Panasonic Holdings Corp
Original Assignee
Matsushita Electric Industrial Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Matsushita Electric Industrial Co Ltd filed Critical Matsushita Electric Industrial Co Ltd
Priority to JP14246592A priority Critical patent/JPH05336510A/en
Publication of JPH05336510A publication Critical patent/JPH05336510A/en
Pending legal-status Critical Current

Links

Landscapes

  • Compression Or Coding Systems Of Tv Signals (AREA)

Abstract

PURPOSE:To simplify a motion vector detecting means and saves the amount of codes for encoding a motion vector by generating a part of the motion vector by interpolation. CONSTITUTION:A motion detecting means 2 compares an input signal 1 with the decoded signal 12 of a previous screen outputted from a memory 11 to detect the motion vector 3. From the motion vector 3, a motion interpolative generating means 13 generates the motion vector 14 of an area which is not detected by the motion detecting means 2. A motion compensating and encoding means 4 compares the input signal 1 with the decoded signal 8 of the contents of a memory 11 corresponding to the motion vector 14 and encodes the difference when the difference is small, but encodes the input signal 1 as it is when not, thereby outputting the encoded signal 5. An encoded signal 6 is also outputted separately from the encoding signal 5 after error correction. A motion compensating and decoding means 7 performs decoding with a decoded signal 8 and the encoded signal 6 to generate a decoded signal 9. The decoded signal 9 is recorded in the memory 11 and used for the encoding and decoding of a next screen.

Description

【発明の詳細な説明】Detailed Description of the Invention

【0001】[0001]

【産業上の利用分野】本発明は、伝送または記録するた
めに画像信号のデータ量を削減するために使用される高
能率符号化手法の一つである動き補償符号化装置と動き
補償復号化装置に関するものである。
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a motion compensation coding apparatus and motion compensation decoding which is one of high efficiency coding techniques used for reducing the data amount of an image signal to be transmitted or recorded. It relates to the device.

【0002】[0002]

【従来の技術】従来の動き補償符号化装置のブロック図
を図6に示す。同図において、1は画像の入力信号、2
は動き検出手段、3は動きベクトル、4は動き補償符号
化手段、5,6は符号化信号、7は動き補償復号化手
段、8は前画面の復号化信号、9は復号化信号、11は
メモリ、12は前画面の復号化信号である。
2. Description of the Related Art A block diagram of a conventional motion compensation coding apparatus is shown in FIG. In the figure, 1 is an image input signal, 2
Is a motion detecting means, 3 is a motion vector, 4 is motion compensation coding means, 5 and 6 are coded signals, 7 is motion compensation decoding means, 8 is a decoded signal of the previous screen, 9 is a decoded signal, 11 Is a memory, and 12 is a decoded signal of the previous screen.

【0003】以上のように構成された従来の動き補償符
号化装置について、以下その動作を説明する。動き検出
手段2では入力信号1とメモリ11の出力である前画面
の復号化信号12を比較し、動きベクトル3を検出す
る。動き検出の具体的な手段としては、例えば、入力信
号1と前画面の復号化信号12とマッチングを取り、マ
ッチング誤差が最小値なる相対位置を動きベクトルとす
る手段がある。動き補償符号化手段4では入力信号1と
動きベクトル3に対応するメモリ11の内容である復号
化信号8を比較し、その差が小さければ差を符号化し、
そうでなければ入力信号1をそのまま符号化して、符号
化信号5を出力する。なお、復号化の際には伝送路や誤
り訂正を除いた圧縮のみが行なわれた符号化信号が必要
であり、符号化信号6として出力する。動き補償復号化
手段7では復号化信号8と符号化信号6から復号を行な
い、復号化信号9を生成する。復号化信号9はメモリ1
1に記録され、次画面の符号化や復号化に使用される。
The operation of the conventional motion compensation coding apparatus configured as described above will be described below. The motion detecting means 2 compares the input signal 1 and the decoded signal 12 of the previous screen output from the memory 11 to detect the motion vector 3. As a concrete means for detecting the motion, there is, for example, means for matching the input signal 1 and the decoded signal 12 of the previous screen and setting the relative position where the matching error is the minimum value as the motion vector. In the motion compensation coding means 4, the input signal 1 is compared with the decoded signal 8 which is the content of the memory 11 corresponding to the motion vector 3, and if the difference is small, the difference is coded,
Otherwise, the input signal 1 is encoded as it is, and the encoded signal 5 is output. It is to be noted that, when decoding, a coded signal that is compressed only by removing the transmission line and error correction is necessary, and is output as a coded signal 6. The motion compensation decoding means 7 decodes the decoded signal 8 and the coded signal 6 to generate a decoded signal 9. Decoded signal 9 is in memory 1
1 and is used for encoding and decoding of the next screen.

【0004】次に、従来の動き補償復号化装置のブロッ
ク図を図7に示す。同図に於いて、21は符号化信号、
22は動き復号手段、23は動きベクトル、24は動き
補償復号化手段、25は復号化信号、26はメモリ、2
8は前画面の復号化信号である。
Next, a block diagram of a conventional motion compensation decoding apparatus is shown in FIG. In the figure, 21 is an encoded signal,
22 is a motion decoding means, 23 is a motion vector, 24 is a motion compensation decoding means, 25 is a decoded signal, 26 is a memory, 2
Reference numeral 8 is a decoded signal of the previous screen.

【0005】以上のように構成された従来の動き補償復
号化装置について、以下その動作を説明する。動き復号
手段22は符号化信号21から動きベクトル23を復号
化し、動きベクトル23を動き補償復号化手段24に送
信する。動き補償復号化手段24ではメモリ26から動
きベクトル23に対応する前画面の復号化信号28を用
いて符号化信号21を復号化する。復号化信号25はメ
モリ26に記憶され、次画面の復号化で使用される。
The operation of the conventional motion compensation decoding apparatus configured as described above will be described below. The motion decoding means 22 decodes the motion vector 23 from the encoded signal 21 and sends the motion vector 23 to the motion compensation decoding means 24. The motion compensation decoding means 24 decodes the coded signal 21 from the memory 26 using the decoded signal 28 of the previous screen corresponding to the motion vector 23. The decoded signal 25 is stored in the memory 26 and used for decoding the next screen.

【0006】[0006]

【発明が解決しようとする課題】しかしながら、上記の
様な構成においては、符号化単位毎に動きベクトルを符
号化することが必要で、動きベクトルを符号化するため
の符号量が多くなり、符号化効率が低下する。更に、動
きベクトル検出手段は大規模なハードウェアが必要であ
り、動きベクトルの検出回数を減少させることによりハ
ードウェア規模を小さくすることが可能である。
However, in the above-mentioned configuration, it is necessary to code the motion vector for each coding unit, and the code amount for coding the motion vector increases, and Efficiency decreases. Further, the motion vector detecting means requires a large-scale hardware, and the hardware scale can be reduced by reducing the number of times the motion vector is detected.

【0007】かかる点に鑑み、本発明は符号化に必要な
動きベクトルを減少してハードウァア規模を低減し、従
来の動きベクトル符号化手法よりも符号化効率を高める
動き補償符号化装置と、その符号化信号を復号化する動
き補償復号化装置を提供することを目的とする。
In view of such a point, the present invention reduces a motion vector required for encoding to reduce a hardware size, and a motion compensation encoding device which enhances encoding efficiency as compared with a conventional motion vector encoding method, and An object of the present invention is to provide a motion compensation decoding device that decodes a coded signal.

【0008】[0008]

【課題を解決するための手段】上記課題を解決するため
に、本発明の動き補償符号化装置は、現画面入力信号と
前画面復号化信号から動きベクトルを検出する動きベク
トル検出手段と、前記動きベクトル検出手段で検出した
動きベクトルから現画面の他の位置の動きベクトルを生
成する動きベクトル生成手段と、前記動きベクトル生成
手段で生成された動きベクトルに対応する前画面復号化
信号を利用して現画面入力信号を符号化する動き補償符
号化手段と、前記動き補償符号化手段で使用した前画面
復号化信号を利用して前記動き補償符号化手段で符号化
した信号を復号化する動き補償復号化手段と、前記動き
補償復号化手段で復号化された信号を次画面で使用する
ために記憶するメモリで構成される。
In order to solve the above problems, a motion compensation coding apparatus of the present invention comprises a motion vector detecting means for detecting a motion vector from a current screen input signal and a previous screen decoded signal, and Utilizing a motion vector generating means for generating a motion vector at another position on the current screen from the motion vector detected by the motion vector detecting means, and a previous screen decoded signal corresponding to the motion vector generated by the motion vector generating means. Motion compensation encoding means for encoding the current screen input signal, and motion for decoding the signal encoded by the motion compensation encoding means using the previous screen decoded signal used in the motion compensation encoding means. It comprises a compensation decoding means and a memory for storing the signal decoded by the motion compensation decoding means for use in the next screen.

【0009】また、本発明の動き補償符号化装置は、現
画面入力信号と前画面復号化信号から動きベクトルを検
出する動きベクトル検出手段と、符号化された動きベク
トルから現画面の他の位置の動きベクトルを生成する動
きベクトル生成手段と、前記動きベクトル検出手段およ
び前記動きベクトル生成手段で生成された動きベクトル
に対応する前画面復号化信号を利用して現画面入力信号
を符号化する動き補償符号化手段と、前記動き補償符号
化手段で使用した前画面復号化信号を利用して前記動き
補償符号化手段で符号化した信号を復号化する動き補償
復号化手段と、前記動き補償復号化手段で復号化された
信号を次画面で使用するために記憶するメモリで構成さ
れる。
Further, the motion compensation coding apparatus of the present invention comprises a motion vector detecting means for detecting a motion vector from the current screen input signal and the previous screen decoded signal, and another position of the current screen from the coded motion vector. Motion vector generating means for generating the motion vector of the current screen, and a motion for encoding the current screen input signal by using the previous screen decoded signal corresponding to the motion vector generated by the motion vector detecting means and the motion vector generating means. Compensation coding means, motion compensation decoding means for decoding the signal coded by the motion compensation coding means using the previous screen decoded signal used in the motion compensation coding means, and the motion compensation decoding It is composed of a memory for storing the signal decoded by the converting means for use in the next screen.

【0010】また、本発明の動き補償復号化装置は、符
号化信号から動きベクトルを復号化する動き復号化手段
と、前記符号化信号と前記動き復号化手段と前画面復号
化信号から動き補償復号化して前記メモリに記憶し動き
補償復号化手段から構成される。
Also, the motion compensation decoding apparatus of the present invention comprises motion decoding means for decoding a motion vector from a coded signal, motion compensation from the coded signal, the motion decoding means, and the previous screen decoded signal. The data is decoded and stored in the memory, and comprises motion compensation decoding means.

【0011】[0011]

【作用】本発明の第1の動き補償符号化装置に於いて
は、現画面入力信号とメモリから読み出された前画面の
復号化信号を動き検出手段で比較し、動きベクトルを検
出する。この動き検出は画面内の一部の領域に対して行
われる。動き補間生成手段では動きベクトル検出が行わ
れない領域に対して近傍位置の動きベクトルから補間に
よって動きベクトルを生成する。前記動きベクトル検出
手段で検出された動きベクトルが符号化されていれば、
動き補間生成手段で生成した動きベクトルは動き補償復
号化装置でも同様に近傍の動きベクトルから生成可能で
あり、従って、前記動きベクトル検出手段で検出した動
きベクトルのみ付加情報として符号化すればよい。この
ようにして全ての領域に対する動きベクトルが生成され
た後、動き補償符号化手段で動き補償符号化が行われ
る。なお、符号化信号を動き補償符号化手段で復号化
し、前記メモリに記録して次画面の復号化信号として使
用する。
In the first motion compensation coding apparatus of the present invention, the motion detection means compares the input signal of the current screen with the decoded signal of the previous screen read from the memory to detect the motion vector. This motion detection is performed on a partial area within the screen. The motion interpolation generation means generates a motion vector by interpolation from a motion vector at a position near a region where motion vector detection is not performed. If the motion vector detected by the motion vector detecting means is encoded,
The motion vector generated by the motion interpolation generating means can also be generated from the nearby motion vector in the motion compensation decoding device as well. Therefore, only the motion vector detected by the motion vector detecting means needs to be encoded as additional information. After the motion vectors for all areas are generated in this way, the motion compensation coding means performs motion compensation coding. The coded signal is decoded by the motion compensation coding means, recorded in the memory and used as a decoded signal for the next screen.

【0012】本発明の第2の動き補償符号化装置に於い
ては、動き補間生成手段で復号化した動きベクトルから
動きベクトルを補間により生成することが第1の発明と
の差異である。その結果、動きベクトルの補間生成手段
は第1の発明よりも複雑になるが、復号化された動きベ
クトルのみを用いるので、補間に必要な動きベクトルが
符号化されるかどうかを判定する必要がない。
In the second motion compensation coding apparatus of the present invention, the difference from the first invention is that the motion vector is generated by interpolation from the motion vector decoded by the motion interpolation generating means. As a result, the motion vector interpolation generating means becomes more complicated than that of the first invention, but since only the decoded motion vector is used, it is necessary to judge whether or not the motion vector necessary for interpolation is coded. Absent.

【0013】本発明の動き補償復号化手段では符号化信
号から動き復号化手段で動きベクトルを復号化し、動き
補間生成手段で符号化されていない領域の動きベクトル
を補間生成する。動き補償復号化手段は、前記符号化信
号と前記動き補間生成手段で生成された動きベクトルと
メモリから読みだされた前画面の復号化信号を用いて復
号化を行う。
In the motion compensation decoding means of the present invention, the motion decoding means decodes the motion vector from the coded signal, and the motion interpolation generation means interpolates and generates the motion vector of the uncoded area. The motion compensation decoding means performs decoding using the coded signal, the motion vector generated by the motion interpolation generation means, and the decoded signal of the previous screen read from the memory.

【0014】動きベクトルは隣接する領域間で強い相関
があり、上記のように動きベクトルを補間生成しても、
直接動き検出手段で検出した動きベクトルとの差は殆ど
なく、その結果、動き補償された信号の符号化効率も劣
化しない。従って、以上の様にして、全ての領域の動き
ベクトルを符号化することなく、動き補償符号化および
動き補償復号化が行え、従来の動き補償符号化装置およ
び動き補償復号化装置よりも動きベクトル符号化に必要
な付加情報量を削減すると共に、動きベクトル検出に必
要な演算量を削減することが出来る。
The motion vector has a strong correlation between adjacent regions, and even if the motion vector is interpolated and generated as described above,
There is almost no difference from the motion vector detected by the direct motion detection means, and as a result, the coding efficiency of the motion-compensated signal does not deteriorate. Therefore, as described above, the motion compensation coding and the motion compensation decoding can be performed without coding the motion vectors of all the regions, and the motion vector can be calculated more than the conventional motion compensation coding apparatus and the motion compensation decoding apparatus. It is possible to reduce the amount of additional information required for encoding and reduce the amount of calculation required for motion vector detection.

【0015】[0015]

【実施例】図1は本発明の第1の実施例である動き補償
符号化装置のブロック図である。同図において、1は画
像の入力信号、2は動き検出手段、3は動きベクトル、
13は動き補間生成手段、14は動きベクトル、4は動
き補償符号化手段、5,6は符号化信号、7は動き補償
復号化手段、8は前画面の復号化信号、9は復号化信
号、11はメモリ、12は前画面の復号化信号である。
1 is a block diagram of a motion compensation coding apparatus according to a first embodiment of the present invention. In the figure, 1 is an image input signal, 2 is a motion detecting means, 3 is a motion vector,
13 is a motion interpolation generating means, 14 is a motion vector, 4 is motion compensation coding means, 5 and 6 are coded signals, 7 is motion compensation decoding means, 8 is a decoded signal of the previous screen, and 9 is a decoded signal. , 11 is a memory, and 12 is a decoded signal of the previous screen.

【0016】以上のように構成された従来の動き補償符
号化装置について、以下その動作を説明する。動き検出
手段2では入力信号1とメモリ11の出力である前画面
の復号化信号12を比較し、動きベクトル3を検出す
る。動きベクトル検出は領域(通常は矩形のブロック)
単位で行われるが、この動きベクトル検出は全ての領域
ではなく一部の領域で行われる。従って、全ての領域に
ついて動きベクトルを検出する場合と比較して演算回数
が大幅に減少する。動きベクトル3から動き補間生成手
段13によって先の動き検出手段2で検出しない領域の
動きベクトル14が生成される。なお、動き検出手段2
で検出された動きベクトル3も動きベクトル14に含ま
れている。例えば、動き検出手段2で図2に示すように
動きベクトルViとVi+3が検出されたとする。動き
補間生成手段13は動きベクトルViとVi+3から補
間により動きベクトルVi+1とVi+2を生成し、動
きベクトルVi,Vi+1,Vi+2とVi+3が動き
ベクトル14として出力する。画像信号では隣接する領
域の動きベクトルは相関が強く、このように補間で動き
ベクトルを生成しても、直接検出した動きベクトルとの
差はわずかであり、動き補償信号の符号化効率は殆ど劣
化しない。動き補償符号化手段4では入力信号1と動き
ベクトル14に対応するメモリ11の内容である復号化
信号8を比較し、その差が小さければ差を符号化し、そ
うでなければ入力信号1をそのまま符号化して、符号化
信号5を出力する。その際に動きベクトル14を付加情
報として符号化するが、動きベクトル14の中で動きベ
クトル3以外のものは動き補償復号化装置において補間
生成可能であり、従って動きベクトル符号化の付加情報
は従来例よりも大幅に削減可能である。なお、動き補償
復号化には符号化信号が必要であり、誤り訂正も行った
符号化信号5と別に符号化信号6も出力する。動き補償
復号化手段7では復号化信号8と符号化信号6から復号
を行ない、復号化信号9を生成する。復号化信号9はメ
モリ11に記録され、次画面の符号化や復号化に使用さ
れる。
The operation of the conventional motion compensation coding apparatus configured as described above will be described below. The motion detecting means 2 compares the input signal 1 and the decoded signal 12 of the previous screen output from the memory 11 to detect the motion vector 3. Motion vector detection area (usually rectangular block)
Although it is performed in units, this motion vector detection is performed in some areas, not in all areas. Therefore, the number of calculations is significantly reduced as compared with the case where motion vectors are detected for all areas. From the motion vector 3, the motion interpolation generating means 13 generates the motion vector 14 of the area which is not detected by the previous motion detecting means 2. The motion detection means 2
The motion vector 3 detected in 3 is also included in the motion vector 14. For example, it is assumed that the motion detecting means 2 detects the motion vectors Vi and Vi + 3 as shown in FIG. The motion interpolation generating means 13 generates motion vectors Vi + 1 and Vi + 2 by interpolation from the motion vectors Vi and Vi + 3, and outputs the motion vectors Vi, Vi + 1, Vi + 2 and Vi + 3 as the motion vector 14. In the image signal, the motion vectors of adjacent regions have a strong correlation, and even if the motion vector is generated by interpolation in this way, the difference from the motion vector detected directly is small, and the coding efficiency of the motion compensation signal is almost degraded. do not do. In the motion compensation coding means 4, the input signal 1 is compared with the decoded signal 8 which is the content of the memory 11 corresponding to the motion vector 14, and if the difference is small, the difference is coded, otherwise the input signal 1 is left as it is. Encode and output the encoded signal 5. At this time, the motion vector 14 is encoded as additional information. However, the motion vector 14 other than the motion vector 3 can be interpolated and generated in the motion compensation decoding device, and therefore the additional information of the motion vector encoding is conventionally used. It can be significantly reduced compared to the example. Note that a coded signal is required for motion compensation decoding, and a coded signal 6 is also output in addition to the coded signal 5 that has undergone error correction. The motion compensation decoding means 7 decodes the decoded signal 8 and the coded signal 6 to generate a decoded signal 9. The decoded signal 9 is recorded in the memory 11 and used for encoding and decoding the next screen.

【0017】以上のように本実施例によれば、従来の動
き補償符号化装置に動き補間生成手段を付加することに
より、動き検出手段2の演算回数を削減し、更に動きベ
クトルを符号化するのに必要な付加情報量を削減するこ
とが可能である。
As described above, according to the present embodiment, by adding the motion interpolation generating means to the conventional motion compensation coding apparatus, the number of calculations of the motion detecting means 2 is reduced and the motion vector is further coded. It is possible to reduce the amount of additional information required for this.

【0018】図3は本発明の第1の実施例である動き補
償符号化装置のブロック図である。同図において、1は
画像の入力信号、2は動き検出手段、3は動きベクト
ル、13は動き補間生成手段、14は動きベクトル、4
は動き補償符号化手段、5,6,15は符号化信号、7
は動き補償復号化手段、8は前画面の復号化信号、9は
復号化信号、11はメモリ、12は前画面の復号化信号
である。
FIG. 3 is a block diagram of a motion compensation coding apparatus according to the first embodiment of the present invention. In the figure, 1 is an image input signal, 2 is motion detection means, 3 is motion vector, 13 is motion interpolation generation means, 14 is motion vector, 4
Is a motion compensation coding means, 5, 6, 15 are coded signals, 7
Is a motion compensation decoding means, 8 is a decoded signal of the previous screen, 9 is a decoded signal, 11 is a memory, and 12 is a decoded signal of the previous screen.

【0019】以上の様に構成された本発明の第2の実施
例について以下その動作を説明する。図3は図1と殆ど
同じであり、動き補間生成手段13の動作が図1の実施
例と異なっている。図1の実施例では動きベクトル3か
ら動きベクトルを補間生成するが、動き補償復号化装置
で動きベクトルを正しく補間するためには動きベクトル
3が符号化されている必要がある。図2の実施例の動き
補償符号化手段4では、動きベクトル3が検出される領
域の符号化が行われるが、その際に動き補償される領域
に対応する動きベクトル3のみを付加情報で符号化し、
動き補償されない領域(画面内符号化)される領域の動
きベクトルは符号化しない。従って、動き補償されない
領域の動きベクトルは動き補間生成手段13では使用す
ることが不可能である。よって、本実施例では、動きベ
クトルを符号化した符号化信号15を復号化し、符号化
済みの動きベクトルから動きベクトル14を補間生成す
る。補間の一例を図4に示す。図4において15は符号
化信号、51aは平均補間手段、51bは前方補間手
段、51cは後方補間手段、52a,52b,52cは
動きベクトル、53はセレクタ、14は動きベクトルで
ある。平均補間手段51aは符号化信号15から動きベ
クトルを復号化し、前後の復号化した動きベクトルから
その間の位置の動きベクトル52aを補間生成する。前
方補間手段51bは符号化信号15から動きベクトルを
復号化し、後位置の復号化した動きベクトルからその前
の位置の動きベクトル52bを補間生成する。後方補間
手段51cは符号化信号15から動きベクトルを復号化
し、前位置の復号化した動きベクトルからその後の位置
の動きベクトル52cを補間生成する。セレクタ53は
このようにして補間生成した動きベクトル52a,52
b,52cの中から所定の基準で1つを選択し、動きベ
クトル14として出力する。選択の基準としては、例え
ば、 ・直前の位置の動きベクトルのみが復号化されていれば
後方補間手段 ・直後の位置の動きベクトルのみが復号化されていれば
前方補間手段 ・上記以外の場合には平均補間手段 と切り替えればよい。
The operation of the second embodiment of the present invention constructed as above will be described below. 3 is almost the same as FIG. 1, and the operation of the motion interpolation generating means 13 is different from that of the embodiment of FIG. In the embodiment of FIG. 1, the motion vector is interpolated and generated from the motion vector 3. However, the motion vector 3 needs to be encoded in order for the motion compensation decoding apparatus to correctly interpolate the motion vector. In the motion compensation coding means 4 of the embodiment shown in FIG. 2, the area in which the motion vector 3 is detected is coded. At this time, only the motion vector 3 corresponding to the area in which the motion is compensated is coded by the additional information. Turned into
The motion vector of the area that is not motion-compensated (intra-picture encoded) is not encoded. Therefore, the motion vector of the area that is not motion-compensated cannot be used by the motion interpolation generating means 13. Therefore, in this embodiment, the coded signal 15 obtained by coding the motion vector is decoded, and the motion vector 14 is interpolated and generated from the coded motion vector. An example of the interpolation is shown in FIG. In FIG. 4, reference numeral 15 is a coded signal, 51a is an average interpolation means, 51b is a forward interpolation means, 51c is a backward interpolation means, 52a, 52b and 52c are motion vectors, 53 is a selector, and 14 is a motion vector. The average interpolation means 51a decodes a motion vector from the encoded signal 15, and interpolates and generates a motion vector 52a at a position between the decoded motion vectors before and after. The front interpolation means 51b decodes the motion vector from the coded signal 15, and interpolates and generates the motion vector 52b at the previous position from the decoded motion vector at the rear position. The backward interpolation means 51c decodes a motion vector from the encoded signal 15, and interpolates and generates a motion vector 52c at a subsequent position from the decoded motion vector at the previous position. The selector 53 uses the interpolation-generated motion vectors 52a and 52a.
One is selected from b and 52c based on a predetermined reference and is output as the motion vector 14. The selection criteria include, for example: backward interpolation means if only the motion vector at the immediately preceding position has been decoded, forward interpolation means if only the motion vector at the immediately succeeding position has been decoded, Should be switched to the mean interpolation means.

【0020】以上のように本実施例によれば、従来の動
き補償符号化装置に動き補間生成手段を付加することに
より、動き検出手段2の演算回数を削減し、更に動きベ
クトルを符号化するのに必要な付加情報量を削減するこ
とが可能である。又、第1の実施例と異なり復号化済み
の動きベクトルのみから動きベクトル14を補間生成す
るので、動きベクトル3を全て符号化することは不要で
ある。
As described above, according to this embodiment, by adding the motion interpolation generating means to the conventional motion compensation coding apparatus, the number of calculations of the motion detecting means 2 is reduced and the motion vector is further coded. It is possible to reduce the amount of additional information required for this. Further, unlike the first embodiment, since the motion vector 14 is generated by interpolation from only the decoded motion vector, it is not necessary to code all the motion vectors 3.

【0021】図5は本発明の第3の実施例である動き補
償復号化装置のブロック図である。同図において、21
は符号化信号、22は動き復号手段、23は動きベクト
ル、30は動き補間生成手段、31は動きベクトル、2
4は動き補償復号化手段、25は復号化信号、26はメ
モリ、28は前画面の復号化信号である。
FIG. 5 is a block diagram of a motion compensation decoding apparatus according to the third embodiment of the present invention. In the figure, 21
Is a coded signal, 22 is a motion decoding means, 23 is a motion vector, 30 is a motion interpolation generating means, 31 is a motion vector, 2
4 is a motion compensation decoding means, 25 is a decoded signal, 26 is a memory, and 28 is a decoded signal of the previous screen.

【0022】以上のように構成された従来の動き補償復
号化装置について、以下その動作を説明する。動き復号
手段22は符号化信号21から動きベクトル23を復号
化し、動きベクトル23を動き補間生成手段30に送信
する。動き補間生成手段30はその位置の動きベクトル
23が存在する場合には動きベクトル23を動きベクト
ル31として出力し、存在しなければ補間により動きベ
クトル23から動きベクトル31を生成する。動き補償
復号化手段24ではメモリ26から動きベクトル31に
対応する前画面の復号化信号28を用いて符号化信号2
1を復号化する。復号化信号25はメモリ26に記憶さ
れ、次画面の復号化で使用される。
The operation of the conventional motion compensation decoding apparatus configured as described above will be described below. The motion decoding means 22 decodes the motion vector 23 from the encoded signal 21 and sends the motion vector 23 to the motion interpolation generating means 30. The motion interpolation generation means 30 outputs the motion vector 23 as the motion vector 31 when the motion vector 23 at that position exists, and when not present, generates the motion vector 31 from the motion vector 23 by interpolation. The motion compensation decoding means 24 uses the decoded signal 28 of the previous screen corresponding to the motion vector 31 from the memory 26 to generate the encoded signal 2.
Decode 1 The decoded signal 25 is stored in the memory 26 and used for decoding the next screen.

【0023】以上のように本実施例によれば、動き補間
生成手段30を動き復号化手段22と動き補償復号化手
段24の間に設置することにより、本発明の動き補償符
号化手段で符号化された信号を首尾良く復号化すること
ができる。
As described above, according to the present embodiment, the motion interpolation generating means 30 is installed between the motion decoding means 22 and the motion compensation decoding means 24, so that the motion compensation coding means of the present invention performs coding. The encoded signal can be successfully decoded.

【0024】[0024]

【発明の効果】以上説明したように、本発明によれば動
きベクトル検出に必要な演算量を削減し、且つ動きベク
トルを符号化するのに必要な付加情報量を削減できるの
で、その実用的効果は大きい。
As described above, according to the present invention, the amount of calculation required for motion vector detection can be reduced and the amount of additional information required for coding a motion vector can be reduced. The effect is great.

【0025】なお、図4において補間手段として平均補
間、前方補間と後方補間の3通りから1通りを選択した
が、この3通りの中の一部のみを使用したり、他の補間
手法を用いても良い。
In FIG. 4, average interpolation, forward interpolation and backward interpolation are selected as one interpolation method in FIG. 4, but only a part of these three methods is used or another interpolation method is used. May be.

【図面の簡単な説明】[Brief description of drawings]

【図1】本発明の第1の実施例である動き補償符号化装
置のブロック図
FIG. 1 is a block diagram of a motion compensation coding apparatus according to a first embodiment of the present invention.

【図2】動きベクトルの補間の説明図FIG. 2 is an explanatory diagram of motion vector interpolation.

【図3】本発明の第2の実施例である動き補償符号化装
置のブロック図
FIG. 3 is a block diagram of a motion compensation coding apparatus according to a second embodiment of the present invention.

【図4】動きベクトルの補間の説明図FIG. 4 is an explanatory diagram of motion vector interpolation.

【図5】本発明の第3の実施例である動き補償復号化装
置のブロック図
FIG. 5 is a block diagram of a motion compensation decoding apparatus that is a third embodiment of the present invention.

【図6】従来の動き補償符号化装置のブロック図FIG. 6 is a block diagram of a conventional motion compensation encoding device.

【図7】従来の動き補償復号化装置のブロック図FIG. 7 is a block diagram of a conventional motion compensation decoding device.

【符号の説明】 2 動き検出手段 4 動き補償符号化手段 7 動き補償復号化手段 11,26 メモリ 13,30 動き補間生成手段 22 動き復号手段 24 動き補償復号化手段[Description of Codes] 2 motion detection means 4 motion compensation coding means 7 motion compensation decoding means 11, 26 memory 13, 30 motion interpolation generation means 22 motion decoding means 24 motion compensation decoding means

Claims (3)

【特許請求の範囲】[Claims] 【請求項1】現画面入力信号と前画面復号化信号から動
きベクトルを検出する動きベクトル検出手段と、前記動
きベクトル検出手段で検出した動きベクトルから現画面
の他の位置の動きベクトルを生成する動きベクトル生成
手段と、前記動きベクトル生成手段で生成された動きベ
クトルに対応する前画面復号化信号を利用して現画面入
力信号を符号化する動き補償符号化手段と、前記動き補
償符号化手段で使用した前画面復号化信号を利用して前
記動き補償符号化手段で符号化した信号を復号化する動
き補償復号化手段と、前記動き補償復号化手段で復号化
された信号を次画面で使用するために記憶するメモリで
構成される動き補償符号化装置。
1. A motion vector detecting means for detecting a motion vector from a current screen input signal and a previous screen decoded signal, and a motion vector at another position of the current screen from the motion vector detected by the motion vector detecting means. Motion vector generation means, motion compensation coding means for coding the current screen input signal using the previous screen decoded signal corresponding to the motion vector generated by the motion vector generation means, and the motion compensation coding means The motion compensation decoding means for decoding the signal encoded by the motion compensation encoding means using the previous screen decoded signal used in the above, and the signal decoded by the motion compensation decoding means on the next screen. A motion compensation coder comprising a memory for storing for use.
【請求項2】現画面入力信号と前画面復号化信号から動
きベクトルを検出する動きベクトル検出手段と、符号化
された動きベクトルから現画面の他の位置の動きベクト
ルを生成する動きベクトル生成手段と、前記動きベクト
ル検出手段および前記動きベクトル生成手段で生成され
た動きベクトルに対応する前画面復号化信号を利用して
現画面入力信号を符号化する動き補償符号化手段と、前
記動き補償符号化手段で使用した前画面復号化信号を利
用して前記動き補償符号化手段で符号化した信号を復号
化する動き補償復号化手段と、前記動き補償復号化手段
で復号化された信号を次画面で使用するために記憶する
メモリで構成される動き補償符号化装置。
2. A motion vector detecting means for detecting a motion vector from a current screen input signal and a previous screen decoded signal, and a motion vector generating means for generating a motion vector at another position of the current screen from the encoded motion vector. A motion compensation coding means for coding a current screen input signal using a previous screen decoded signal corresponding to the motion vector generated by the motion vector detection means and the motion vector generation means; and the motion compensation code. Next, a motion compensation decoding means for decoding the signal encoded by the motion compensation encoding means using the front screen decoded signal used by the encoding means, and a signal decoded by the motion compensation decoding means are A motion compensation coding device consisting of a memory that stores for use on a screen.
【請求項3】符号化信号から動きベクトルを復号化する
動き復号化手段と、前記符号化信号と前記動き復号化手
段と前画面復号化信号から動き補償復号化して前記メモ
リに記憶し動き補償復号化手段構成される動き補償復号
化装置。
3. A motion decoding means for decoding a motion vector from a coded signal, a motion compensation decoding from the coded signal, the motion decoding means, and a previous screen decoded signal, which is stored in the memory and motion compensated. A motion-compensated decoding device comprising decoding means.
JP14246592A 1992-06-03 1992-06-03 Motion compensation encoding device and decoding device Pending JPH05336510A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP14246592A JPH05336510A (en) 1992-06-03 1992-06-03 Motion compensation encoding device and decoding device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP14246592A JPH05336510A (en) 1992-06-03 1992-06-03 Motion compensation encoding device and decoding device

Publications (1)

Publication Number Publication Date
JPH05336510A true JPH05336510A (en) 1993-12-17

Family

ID=15315952

Family Applications (1)

Application Number Title Priority Date Filing Date
JP14246592A Pending JPH05336510A (en) 1992-06-03 1992-06-03 Motion compensation encoding device and decoding device

Country Status (1)

Country Link
JP (1) JPH05336510A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2004008775A1 (en) * 2002-07-15 2004-01-22 Hitachi, Ltd. Moving picture encoding method and decoding method
JP2008283231A (en) * 2007-05-08 2008-11-20 Hitachi Ltd Image conversion device

Cited By (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8325816B2 (en) 2002-07-15 2012-12-04 Hitachi Consumer Electronics Co., Ltd. Moving picture encoding method and decoding method
US7936822B2 (en) 2002-07-15 2011-05-03 Hitachi Consumer Electronics Co., Ltd. Moving picture encoding method and decoding method
US8340190B2 (en) 2002-07-15 2012-12-25 Hitachi Consumer Electronics Co., Ltd. Moving picture encoding method and decoding method
US8571108B2 (en) 2002-07-15 2013-10-29 Hitachi Consumer Electronics Co., Ltd. Moving picture encoding method and decoding method
US7936821B2 (en) 2002-07-15 2011-05-03 Hitachi Consumer Electronics Co., Ltd. Moving picture encoding method and decoding method
US7970058B2 (en) 2002-07-15 2011-06-28 Hitachi Consumer Electronics Co., Ltd. Moving picture encoding method and decoding method
US8036272B2 (en) 2002-07-15 2011-10-11 Hitachi Consumer Electronics Co., Ltd. Moving picture encoding method and decoding method
US8320459B2 (en) 2002-07-15 2012-11-27 Hitachi Consumer Electronics Co., Ltd. Moving picture encoding method and decoding method
US10349062B2 (en) 2002-07-15 2019-07-09 Maxell, Ltd. Moving picture encoding method and decoding method considering motion vectors of blocks adjacent to target block
WO2004008775A1 (en) * 2002-07-15 2004-01-22 Hitachi, Ltd. Moving picture encoding method and decoding method
US7936823B2 (en) 2002-07-15 2011-05-03 Hitach Consumer Electronics Co., Ltd. Moving picture encoding method and decoding method
US8571107B2 (en) 2002-07-15 2013-10-29 Hitachi Consumer Electronics Co., Ltd. Moving picture encoding method and decoding method
US8837594B2 (en) 2002-07-15 2014-09-16 Hitachi Consumer Electronics Co., Ltd. Moving picture encoding method and decoding method considering motion vectors of blocks adjacent to target block
US9832473B2 (en) 2002-07-15 2017-11-28 Hitachi Maxell, Ltd. Moving picture encoding method and decoding method with motion vectors of blocks adjacent to target blocks and motion vectors at a same position to target blocks in other frames
US10104383B2 (en) 2002-07-15 2018-10-16 Maxell, Ltd. Moving picture encoding method and decoding method considering motion vectors of blocks adjacent to target block
US10110906B2 (en) 2002-07-15 2018-10-23 Maxell, Ltd. Moving picture encoding method and decoding method considering motion vectors of blocks adjacent to target block
US10327004B2 (en) 2002-07-15 2019-06-18 Maxell, Ltd. Moving picture encoding method and decoding method considering motion vectors of blocks adjacent to target block
JP2008283231A (en) * 2007-05-08 2008-11-20 Hitachi Ltd Image conversion device

Similar Documents

Publication Publication Date Title
US4866510A (en) Digital video encoder
JPH03133297A (en) Video signal coding circuit
EP0357386A2 (en) Image encoding apparatus
EP0529587B1 (en) Image encoding apparatus
JPH0779441A (en) Decoding method for picture data
JPH05336510A (en) Motion compensation encoding device and decoding device
JP2875549B2 (en) Video encoding device and video decoding device
JP2939996B2 (en) Adaptive Frame Interpolation for Inter-frame Prediction Decoder with Motion Compensation
JP3226637B2 (en) Coding device and decoding device for predictive coding method
JPH02171092A (en) Inter-frame coding decoding system
JP2894334B2 (en) Image signal error correction circuit
JPH05207442A (en) Coder for animation signal
JPH05308662A (en) High efficiency encoder
US5206725A (en) Method and apparatus for coding/decoding image signal providing accurate determination of an image contour and efficient compression coding
JPH11136683A (en) Video signal coder
JPH0368597B2 (en)
JPH05236454A (en) Method and device for detecting motion vector for encoding image
JP3469597B2 (en) Decoding device for block transform code
JP2831509B2 (en) High efficiency coding apparatus and high efficiency coding method
JPH0865679A (en) Moving image encoding device and moving image decoding device
JPH01108861A (en) Picture coder
JP2953351B2 (en) Predictive coding device for television signals
JPH08317403A (en) Device and method for image processing
JPH06245084A (en) Processor for picture data
JPH10150664A (en) Encoding device and decoding device for video signal