CN104661035A - Compression method and system for local feature descriptor of video and video compression method - Google Patents
Compression method and system for local feature descriptor of video and video compression method Download PDFInfo
- Publication number
- CN104661035A CN104661035A CN201510073614.7A CN201510073614A CN104661035A CN 104661035 A CN104661035 A CN 104661035A CN 201510073614 A CN201510073614 A CN 201510073614A CN 104661035 A CN104661035 A CN 104661035A
- Authority
- CN
- China
- Prior art keywords
- video
- coefficient
- local feature
- descriptor
- quantization
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 64
- 238000007906 compression Methods 0.000 title claims abstract description 56
- 230000006835 compression Effects 0.000 title claims abstract description 53
- 238000013139 quantization Methods 0.000 claims abstract description 69
- 230000009466 transformation Effects 0.000 claims description 23
- 238000004364 calculation method Methods 0.000 claims description 12
- 230000001131 transforming effect Effects 0.000 claims description 5
- 230000005540 biological transmission Effects 0.000 abstract description 8
- 239000011159 matrix material Substances 0.000 description 6
- 238000006243 chemical reaction Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000002349 favourable effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention discloses a compression method and system for a local feature descriptor of a video and a video compression method. The compression method for the local feature descriptor of the video comprises the following steps: performing inter-frame prediction, quantization and entropy coding on the combined video content, calculating a residual coefficient of a prediction signal, quantizing the residual coefficient to obtain a quantization coefficient, performing entropy coding on the quantization coefficient, outputting a code stream formed after entropy coding, and finishing compression of the local feature descriptor of the video. The local feature descriptor of the video is compressed through the compression method and system, compressed video data is compactly expressed on the basis of the combined video content, and high compression ratio is achieved, so that the transmission rate of the video data is increased, the storage efficiency and the compression efficiency of the video data are improved, and the retrieval efficiency of the video data is improved.
Description
Technical Field
The invention belongs to the technical field of data processing, and particularly relates to a compression method, a compression system and a video compression method for a video local feature descriptor.
Background
With the wide application of the internet, the data of the video data is increasingly huge in the transmission and storage process of the network. The compression of video data is particularly important. The local feature descriptors are widely applied to the fields of image/video data processing and computer vision, and the compression of the descriptors is also a key technology for improving video data applications such as video data retrieval. For example, for a 128-dimensional SIFT local feature descriptor, each dimension is quantized to 8bits, and a frame of image contains 1000 SIFT feature points, so that a 300-frame video contains 300 × 1000 × 128 × 8bits 292.97Mb of SIFT feature data, which is very expensive to store and transmit and is not acceptable in practical applications.
In the prior art of video data processing, descriptors are mostly compressed separately, which limits the compression ratio of descriptors, is not favorable for improving data retrieval efficiency, and also hinders the development of video retrieval technology.
Disclosure of Invention
Embodiments of the present invention provide a method and a system for compressing a video local feature descriptor, and a video compression method, so as to improve compression efficiency of the video local feature descriptor, thereby improving transmission rate and storage efficiency of video compressed data.
According to an aspect of the present invention, there is provided a method for compressing a video local feature descriptor, the method comprising the steps of:
performing inter-frame prediction on the current local feature descriptor by combining video content to obtain a prediction signal;
calculating residual coefficients of the prediction signal;
quantizing the residual error coefficient to obtain a quantized coefficient;
and entropy coding the quantization coefficient and outputting a code stream.
In the above scheme, after the residual coefficients are calculated and before quantized coefficients are obtained, the method further includes:
transforming the residual error coefficient to obtain a transformation coefficient;
and quantizing the residual error coefficient to obtain a quantized coefficient, and further quantizing the transform coefficient to obtain a quantized coefficient.
In the above scheme, the method further comprises: and carrying out inverse quantization and inverse transformation on the quantization coefficient to obtain a reconstruction descriptor, and storing the reconstruction descriptor.
In the foregoing solution, the inter-frame prediction of the joint video content on the current local feature descriptor to obtain a prediction signal further includes: and finding a reconstruction descriptor which is closest to the current local feature descriptor in an encoded previous frame of the frame where the current local feature descriptor is located as a prediction signal.
In the above scheme, the quantization is scalar quantization or vector quantization.
According to another aspect of the present invention, there is also provided a system for compressing a video local feature descriptor, the system including: prediction module, residual coefficient calculation module, quantization module and coding module
The prediction module is used for carrying out inter-frame prediction on the current local feature descriptor by combining video content to obtain a prediction signal;
the residual coefficient calculation module is connected with the prediction module and used for calculating the residual coefficient of the prediction signal;
the quantization module is connected with the residual error coefficient calculation module and is used for quantizing the residual error coefficient to obtain a quantization coefficient;
and the coding module is connected with the quantization module and used for entropy coding the quantization coefficient and outputting a code stream.
In the above scheme, the system further includes a transform module, connected to the residual coefficient calculation module and the quantization module, and configured to transform the residual coefficient to obtain a transform coefficient;
the quantization module is further configured to quantize the transform coefficient and obtain a quantized coefficient.
In the above solution, the system further includes: and the reconstruction descriptor storage module is used for carrying out inverse quantization and inverse transformation on the quantization coefficient to obtain a reconstruction descriptor and storing the reconstruction descriptor.
In the above solution, the prediction module is further configured to find a reconstructed descriptor that is closest to the current local feature descriptor in an encoded previous frame of a frame where the current local feature descriptor is located, as the prediction signal.
According to still another aspect of the present invention, there is also provided a video compression method, including:
compressing original video data to obtain a video code stream and obtain video reconstruction data consisting of reconstruction frames;
combining the content of the reconstructed frame in the video reconstruction data, and compressing the local feature descriptors in the video original data by adopting the compression method of the local feature descriptors in any one of claims 1 to 5 to obtain an entropy-encoded descriptor stream;
and outputting the description sub-code stream to the video code stream and outputting the video code stream.
The embodiment of the invention provides a compression method, a compression system and a compression method of a video local feature descriptor, wherein the compression method of the video local feature descriptor comprises prediction, quantization and entropy coding, the prediction is inter-frame prediction of combined video content, then a residual coefficient of a prediction signal is calculated, the residual coefficient is quantized to obtain a quantization coefficient, then the quantization coefficient is entropy coded, a code stream is output, and the compression of the video local feature descriptor is completed. The method of the embodiment of the invention compresses the local feature descriptors of the video, enables the compressed video data to be compactly expressed on the basis of combining the video content, and achieves high compression ratio, thereby improving the transmission rate and the storage efficiency of the video data and simultaneously improving the retrieval efficiency of the video data.
Drawings
FIG. 1 is a flow chart of a method for compressing a video local feature descriptor according to a first embodiment of the present invention;
FIG. 2 is a flow chart of a method for compressing a video local feature descriptor according to a second embodiment of the present invention;
FIG. 3 is a flow chart of a compression method according to a second embodiment of the present invention;
FIG. 4 is a schematic diagram of a compression system of a video local feature descriptor according to a third embodiment of the present invention;
FIG. 5 is a schematic diagram of a compression system of a video local feature descriptor according to a fourth embodiment of the present invention;
fig. 6 is a flowchart illustrating a video compression method according to a fifth embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail with reference to the accompanying drawings in conjunction with the following detailed description. It should be understood that the description is intended to be exemplary only, and is not intended to limit the scope of the present invention. Moreover, in the following description, descriptions of well-known structures and techniques are omitted so as to not unnecessarily obscure the concepts of the present invention.
The embodiment of the invention is based on the wide application of a local feature descriptor in the field of computer vision, compresses the local feature descriptor of a video, applies the reconstructed video content with the reconstructed descriptor content to a compression process, namely combines the video content to guide the compression of the local feature descriptor, and provides a compression method, a compression system and a video compression method of the local feature descriptor of the video. The present invention will be described in detail below with reference to the embodiments and the accompanying drawings.
Fig. 1 is a schematic flow chart of a method for compressing a video local feature descriptor according to a first embodiment of the present invention. The present embodiment includes prediction, quantization, and entropy coding of joint video content for the framework of the compression method of the video local feature descriptor.
As shown in fig. 1, the method for compressing a video local feature descriptor of this embodiment includes the following steps:
and step S101, inter-frame prediction is carried out on the current local feature descriptor by combining the video content to obtain a prediction signal.
In this step, inter-frame prediction is performed on the current local feature descriptor, specifically, a reconstructed descriptor that is closest to the current local feature descriptor is found in an encoded frame of the video and is used as a prediction signal. Video data is made up of frames, each of which has a corresponding local feature descriptor. The frame to which the current local feature descriptor in the compression process belongs, i.e. the current frame. Preferably, the encoded frame of the video is a frame previous to the current frame. The most similar means that the encoded reconstruction descriptor has the most identical feature data with the current descriptor, and may be selected by Sum of Absolute Difference (SAD). This selection of the most significant value can be accomplished by a Programmable Logic Controller (PLC). When searching the closest reconstruction descriptor in the previous frame, the closest reconstruction descriptor can be searched in all data of the previous frame, or the search range can be customized, for example, the search is performed according to the motion vector of the video, and the closest reconstruction descriptor is searched in the range of five nodes closest to the position pointed by the motion vector of the current descriptor.
Particularly, during the inter-frame prediction, the reconstruction descriptor serving as the prediction result of the current local feature descriptor may be defined as needed, such as a reconstruction descriptor at a corresponding position and a reconstruction descriptor of an opposite feature, in addition to the closest reconstruction descriptor in the corresponding range.
Step S102, calculating a residual coefficient of the prediction signal.
Preferably, the prediction residual is calculated in this step by a difference method.
And step S104, quantizing the residual error coefficient to obtain a quantized coefficient.
The quantization method in this step may be scalar quantization or vector quantization.
And step S105, entropy coding is carried out on the quantization coefficient, and a code stream is output.
In the entropy coding in this step, preferably, the coding model adopts a Context-Adaptive Binary entropy coding method (CABAC) commonly used in video coding standards, and finally forms a Binary code stream.
By the method for compressing the video local feature descriptors, the video local feature descriptors are compressed, and compressed video data are compactly represented on the basis of combining video content, so that a high compression ratio is achieved, the transmission rate and the storage efficiency of the video data are improved, and the retrieval efficiency of the video data is improved.
Fig. 2 is a flow chart of a method for compressing a video local feature descriptor according to a second embodiment of the present invention.
As shown in fig. 2, steps S201, S202, and S205 in the present embodiment are basically the same as steps S102, S104, and S105 in the first embodiment, except that in the method for compressing the local feature descriptor in the present embodiment, step S203 is further included after step S202 and before step S204, and the corresponding step S204 is different, specifically:
step S203, transforming the residual error coefficient to obtain a transformation coefficient.
Preferably, the transform mode in this step is DCT transform, KLT transform, or DST transform, thereby obtaining transform coefficients. The transformation may be a one-dimensional coefficient matrix or a two-dimensional coefficient matrix.
Step S204, quantizing the transformation coefficient and obtaining a quantized coefficient. The quantization may be scalar quantization or vector quantization.
Fig. 3 is a flow chart of a compression method according to a second embodiment of the present invention.
The present embodiment takes a 128-dimensional SIFT local feature descriptor of video data as an example to describe the compression method of the present invention in detail.
As shown in fig. 3, the method for compressing a local feature descriptor based on a video according to this embodiment includes the following steps:
step S301, firstly, the video content carried by the original video frame is encoded by a video encoder, so as to obtain the motion vector information of the video.
Step S302, extracting a local feature descriptor from the original video frame, specifically:
suppose that the jth local feature descriptor of the ith frame currently being encoded isWhereinIs the spatial position information of the descriptor,is a local feature descriptor sub-vector.
Step S303, inter-frame prediction is carried out on the extracted local feature descriptors according to the formula (1)
Inter prediction is performed, i.e.:
first, the current position is obtained in a video encoderMotion vector MV (d)i,dx,dy)，di,dx,dyRespectively, an offset value (di) for the time dimension and an offset value (dx, dy) for the two-dimensional spatial coordinates. Is arranged at (i-d)i)thIn a frame withSearch range Ψ (containing K) with coordinates as starting pointΨOne feature point), a prediction signal which is an inter prediction signal is obtained
Step S304, the original signal and the prediction signal are subtracted to obtain a residual error coefficient
Step S305, for residual error coefficientThe conversion is performed as in steps S305A1 to S305C 1:
step S305A1, converting 128-dimensionalDivided into two 8x8 two-dimensional matrices;
step S305B1, respectively using two DCT transformations of 8x8 to act on the two transformation matrixes to obtain two transformed coefficient matrixes;
step S305C1, merging two coefficient matrixes 8x8 into a 128-dimensional transformation coefficient
Alternatively, the residual coefficients are transformed as in steps S305a2 to S305C 2:
step S305A2, converting 128-dimensionalDividing the three-dimensional matrix into 8 two-dimensional matrixes of 4x4 according to 8 different directions, namely arranging 16 elements of each direction into a matrix of 4x4 according to positions;
step S305B2, respectively using 8 DCT transforms of 4x4 to act on the 8 transform matrixes to obtain 8 transformed coefficient matrixes;
step S305C2, combining 8 coefficient matrixes of 4x4 into a 128-dimensional transformation coefficient
Step S305 may further include:
in step S305D, it is determined whether to perform transform by rate-distortion selection, and if the transform is skipped, the residual coefficients are the transform coefficients quantized at this time. When the residual coefficients are directly quantized by skipping the transform, the embodiment of the first embodiment is described.
Step S306, carrying out scalar Quantization on the transformation coefficient through the set Quantization step (Qs)Obtaining quantized coefficientsThe quantization may be vector quantization, and the present embodiment is preferably scalar quantization.
Step S307, quantizing the coefficient by entropy encoderEntropy Coding is carried out, a Coding model adopts a Context self-Adaptive Binary entropy Coding method (CABAC) commonly used in a video Coding standard, a binarization code stream is finally formed, and the binarization processing process in the entropy Coding comprises the following specific steps:
step S307A, outputting a bin representationWhether the coefficient is an all-zero coefficient or not, if so, ending the algorithm;
step S307B, outputting a binary vector, wherein each bin representsWhether the corresponding coefficient in (1) is zero or not;
step S307C, for each non-zero coefficient, first outputting a sign bit, and then outputting a bin indicating whether the absolute value of the coefficient is 1; if not, outputting a bin to indicate whether the absolute value of the coefficient is 2; if not, the exponential golomb code after subtracting 3 from the absolute value of this coefficient is output. And obtaining a reconstruction descriptor after the encoding is finished, and outputting a code stream.
Step S308, quantizing the quantized coefficientAnd carrying out inverse quantization and/or inverse transformation to obtain a reconstruction descriptor, and storing the reconstruction descriptor in a cache.
When entropy encoding is performed in step S307, the transform control flag is also entropy encoded, and the transform process is recorded in the encoding. The conversion control flag here indicates whether or not the conversion in step S305 and the inverse conversion in step S308 are performed. The transformation and the inverse transformation are correspondingly generated, if the transformation is carried out, the inverse transformation is required to be carried out after the inverse quantization when the descriptor is reconstructed; if no transform is performed during compression, then there is no need to perform an inverse transform after inverse quantization when reconstructing the descriptor.
The method for compressing the local feature descriptors based on the video according to the embodiment compresses the local feature descriptors of the video, and enables the compressed video data to be compactly represented on the basis of combining video content, so that a high compression ratio is achieved, the transmission rate and the storage efficiency of the video data are improved, and the retrieval efficiency of the video data is improved.
Fig. 4 is a schematic diagram of a compression system based on a local feature descriptor of a video according to a third embodiment of the present invention.
As shown in fig. 4, the compression system based on the local feature descriptor of the video according to this embodiment includes: a prediction module 11, a residual coefficient calculation module 12, a quantization module 14 and a coding module 15; wherein,
the prediction module 11 is configured to perform inter-frame prediction on the current local feature descriptor in combination with the video content to obtain a prediction signal.
The prediction module 11 performs inter-frame prediction on the current local feature descriptor, specifically, the prediction module 11 finds a reconstructed descriptor that is closest to the current local feature descriptor in an encoded frame of the video as a prediction signal. Video data is made up of frames, each of which has a corresponding local feature descriptor. The frame to which the current local feature descriptor in the compression process belongs, i.e. the current frame. Preferably, the encoded frame of the video is a frame previous to the current frame. The most similar means that the encoded reconstruction descriptor has the most identical feature data with the current descriptor, and may be selected by Sum of Absolute Difference (SAD). This selection of the most significant value can be accomplished by a Programmable Logic Controller (PLC). When the prediction module 11 searches the closest reconstruction descriptor in the previous frame, it may search all data of the previous frame, or may customize a search range, for example, search according to the motion vector of the video, and search the closest reconstruction descriptor in a range of five nodes closest to the position pointed by the motion vector of the current descriptor.
Specifically, when the prediction module 11 performs inter-frame prediction, the reconstruction descriptor serving as the prediction result of the current local feature descriptor may perform other definitions as needed, such as a reconstruction descriptor at a corresponding position and a reconstruction descriptor of an opposite feature, in addition to the closest reconstruction descriptor in the corresponding range.
The residual coefficient calculating module 12 is connected to the predicting module 11, and is configured to calculate a residual coefficient of the predicted signal. Preferably, the residual coefficient of the prediction signal is calculated in this step by a difference method.
The quantization module 14 is connected to the prediction selection module 12, and is configured to receive the residual coefficient output by the prediction module 12, and quantize the residual coefficient to obtain a quantized coefficient, and output the quantized coefficient. The quantization module 14 performs scalar quantization, or vector quantization, on the residual coefficients.
The encoding module 15 is connected to the quantization module 14, and is configured to receive the quantization coefficient output by the quantization module 14, and perform entropy encoding on the quantization coefficient, and output a code stream.
By the video local feature descriptor compression system, the video local feature descriptors are compressed, compressed video data are compactly represented on the basis of combining video content, and a high compression ratio is achieved, so that the transmission rate and the storage efficiency of the video data are improved, and the retrieval efficiency of the video data is improved.
Fig. 5 is a schematic diagram of a compression system based on a local feature descriptor of a video according to a fourth embodiment of the present invention.
As shown in fig. 5, the compressing system of the descriptor according to this embodiment and the compressing system of the local feature descriptor according to the third embodiment include the same prediction module 11, residual coefficient calculation module 12, and encoding module 15, except that the compressing system of this embodiment further includes a transformation module 13 and a reconstruction descriptor storage module 16, and the connection relationship of the quantization module 14 is different, specifically:
the transformation module 13 is connected to the residual coefficient calculation module 12 and the quantization module 14, and is configured to receive the residual coefficient output by the residual coefficient calculation module 12, and transform the residual coefficient to obtain and output a transformation coefficient.
The quantization module 14 is configured to receive the transform coefficient output by the transform module 13, and quantize the transform coefficient to obtain a quantized coefficient.
The reconstruction descriptor storage module 16 is connected to the encoding module 15 and the prediction module 11, and is configured to perform inverse quantization and inverse transformation on the quantization coefficient to obtain a reconstruction descriptor, and store the reconstruction descriptor.
Fig. 6 is a flowchart illustrating a video compression method according to a fifth embodiment of the present invention.
As shown in fig. 6, the video compression method according to the present embodiment includes:
and S400, compressing the video original data to obtain a video code stream and obtain the motion vector information of the video.
Step S401, inter-frame prediction is performed on the current local feature descriptor by combining the video content to obtain a prediction signal.
In this step, inter-frame prediction is performed on the current local feature descriptor, specifically, a reconstructed descriptor that is closest to the current local feature descriptor is found as a prediction signal in an encoded frame of the video according to the motion vector information.
Step S402, calculating a residual coefficient of the prediction signal. Preferably, the residual coefficient of the prediction signal is calculated in this step by a difference method.
Step S403, transforming the residual coefficient to obtain a transform coefficient.
In this step, preferably, the transform mode is a mode in which DCT acts on the two-dimensional matrix to obtain a coefficient matrix, thereby obtaining a transform coefficient.
Step S404, quantizing the transform coefficient after the residual coefficient transform, and obtaining a quantized coefficient. Preferably, the quantization is scalar quantization.
Step S405, entropy coding is carried out on the quantization coefficient, compression of a local feature descriptor in the video original data is completed, and a descriptor code stream after entropy coding is obtained. In the entropy Coding in this step, preferably, the Coding model adopts a Context-Adaptive Binary entropy Coding method (CABAC) commonly used in video Coding standards, and finally forms a Binary description subcode stream.
Step S406, outputting the description sub-code stream to the video code stream itself, and outputting the video code stream.
By the video compression method, the video is compressed, and the local feature descriptors of the video are compressed on the basis of the combined video content, so that the compressed video data can be compactly represented, and a high compression ratio is achieved, thereby improving the transmission rate and the storage efficiency of the video data, and simultaneously improving the retrieval efficiency of the video data.
It will be understood by those skilled in the art that all or part of the steps and modules for implementing the method and system of the present invention may be implemented by hardware, or by hardware programmed with program instructions, which may be stored in a computer readable storage medium such as a disk, a memory, etc.
It is to be understood that the above-described embodiments of the present invention are merely illustrative of or explaining the principles of the invention and are not to be construed as limiting the invention. Therefore, any modification, equivalent replacement, improvement and the like made without departing from the spirit and scope of the present invention should be included in the protection scope of the present invention. Further, it is intended that the appended claims cover all such variations and modifications as fall within the scope and boundaries of the appended claims or the equivalents of such scope and boundaries.
Claims (10)
1. A method for compressing a video local feature descriptor, the method comprising the steps of:
performing inter-frame prediction on the current local feature descriptor by combining video content to obtain a prediction signal;
calculating residual coefficients of the prediction signal;
quantizing the residual error coefficient to obtain a quantized coefficient;
and entropy coding the quantization coefficient and outputting a code stream.
2. The method of claim 1, wherein after the residual coefficients are calculated and before the quantized coefficients are obtained, the method further comprises:
transforming the residual error coefficient to obtain a transformation coefficient;
and quantizing the residual error coefficient to obtain a quantized coefficient, and further quantizing the transform coefficient to obtain a quantized coefficient.
3. The method of compressing a video local feature descriptor according to claim 1, wherein the method further comprises: and carrying out inverse quantization and inverse transformation on the quantization coefficient to obtain a reconstruction descriptor, and storing the reconstruction descriptor.
4. The method of claim 3, wherein the inter-predicting the current local feature descriptor to obtain a prediction signal from the joint video content, further comprises: and finding a reconstruction descriptor which is closest to the current local feature descriptor in an encoded previous frame of the frame where the current local feature descriptor is located as a prediction signal.
5. The method of claim 4, wherein the quantization is scalar quantization or vector quantization.
6. A system for compressing a video local feature descriptor, the system comprising: prediction module, residual coefficient calculation module, quantization module and coding module
The prediction module is used for carrying out inter-frame prediction on the current local feature descriptor by combining video content to obtain a prediction signal;
the residual coefficient calculation module is connected with the prediction module and used for calculating the residual coefficient of the prediction signal;
the quantization module is connected with the residual error coefficient calculation module and is used for quantizing the residual error coefficient to obtain a quantization coefficient;
and the coding module is connected with the quantization module and used for entropy coding the quantization coefficient and outputting a code stream.
7. The system for compressing a video local feature descriptor according to claim 6, further comprising a transform module, connected to the residual coefficient calculation module and the quantization module, for transforming the residual coefficients to obtain transform coefficients;
the quantization module is further configured to quantize the transform coefficient and obtain a quantized coefficient.
8. The system for compressing a video local feature descriptor according to claim 6, wherein said system further comprises: and the reconstruction descriptor storage module is used for carrying out inverse quantization and inverse transformation on the quantization coefficient to obtain a reconstruction descriptor and storing the reconstruction descriptor.
9. The system of claim 8, wherein the prediction module is further configured to find a reconstructed descriptor that is closest to the current local feature descriptor in an encoded previous frame of the frame where the current local feature descriptor is located, as the prediction signal.
10. A method of video compression, the method comprising:
compressing original video data to obtain a video code stream;
combining the content of the video, compressing the local feature descriptors in the original video data by adopting the compression method of the local feature descriptors of the video according to any one of claims 1 to 5 to obtain an entropy-encoded descriptor code stream;
and outputting the description sub-code stream to the video code stream and outputting the video code stream.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510073614.7A CN104661035B (en) | 2015-02-11 | 2015-02-11 | Compression method, system and the video-frequency compression method of video local feature description |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510073614.7A CN104661035B (en) | 2015-02-11 | 2015-02-11 | Compression method, system and the video-frequency compression method of video local feature description |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104661035A true CN104661035A (en) | 2015-05-27 |
| CN104661035B CN104661035B (en) | 2018-05-29 |
Family
ID=53251658
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510073614.7A Active CN104661035B (en) | 2015-02-11 | 2015-02-11 | Compression method, system and the video-frequency compression method of video local feature description |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104661035B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113473154A (en) * | 2021-06-30 | 2021-10-01 | 杭州海康威视数字技术股份有限公司 | Video encoding method, video decoding method, video encoding device, video decoding device and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120229311A1 (en) * | 2011-03-07 | 2012-09-13 | Industrial Technology Research Institute | Device and method for compressing feature descriptor |
| CN102929969A (en) * | 2012-10-15 | 2013-02-13 | 北京师范大学 | Real-time searching and combining technology of mobile end three-dimensional city model based on Internet |
-
2015
- 2015-02-11 CN CN201510073614.7A patent/CN104661035B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120229311A1 (en) * | 2011-03-07 | 2012-09-13 | Industrial Technology Research Institute | Device and method for compressing feature descriptor |
| CN102929969A (en) * | 2012-10-15 | 2013-02-13 | 北京师范大学 | Real-time searching and combining technology of mobile end three-dimensional city model based on Internet |
Non-Patent Citations (1)
| Title |
|---|
| MINA MAKAR等: "Interframe Coding of Feature Descriptors for Mobile Augmented Reality", 《IEEE TRANSACTIONS ON IMAGE PROCESSING》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113473154A (en) * | 2021-06-30 | 2021-10-01 | 杭州海康威视数字技术股份有限公司 | Video encoding method, video decoding method, video encoding device, video decoding device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104661035B (en) | 2018-05-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102835111B (en) | The motion vector of previous block is used as the motion vector of current block, image to be carried out to the method and apparatus of coding/decoding | |
| EP1379000B1 (en) | Signal encoding method and apparatus and decoding method and apparatus | |
| US20180115787A1 (en) | Method for encoding and decoding video signal, and apparatus therefor | |
| CN107743239B (en) | Method and device for encoding and decoding video data | |
| CN103098469A (en) | Method and apparatus for entropy encoding/decoding a transform coefficient | |
| US10368086B2 (en) | Image coding/decoding method, device, and system | |
| KR20110112171A (en) | Method and apparatus for video encoding, and method and apparatus for video decoding using adaptive coefficient scan order | |
| CN101009839A (en) | Method for video encoding or decoding based on orthogonal transform and vector quantization, and apparatus thereof | |
| CN103188494A (en) | Apparatus and method for encoding/decoding depth image by skipping discrete cosine transform | |
| CN105474642A (en) | Re-encoding image sets using frequency-domain differences | |
| JP2020205609A (en) | Image coding and decoding method, image coding and decoding device, and corresponding computer program | |
| US20110090954A1 (en) | Video Codes with Directional Transforms | |
| CN105025298A (en) | A method and device of encoding/decoding an image | |
| CN109076248A (en) | The vector quantization of Video coding is carried out by the code book generated by the training signal selected | |
| CN106576166A (en) | Image encoding and decoding method, image encoding and decoding device and corresponding computer program | |
| CN104661035B (en) | Compression method, system and the video-frequency compression method of video local feature description | |
| CN1848960B (en) | Residual coding in compliance with a video standard using non-standardized vector quantization coder | |
| CN104735459B (en) | Compression method, system and the video-frequency compression method of video local feature description | |
| WO2015068051A2 (en) | Method for encoding and decoding a media signal and apparatus using the same | |
| CN103533351A (en) | Image compression method for multi-quantization table | |
| CN104661034B (en) | The compression method and system of local feature description's based on video | |
| Lu et al. | Image Compression Based on Mean Value Predictive Vector Quantization. | |
| JP2015035788A (en) | Image encoder, image encoding method, and program | |
| CN113473154B (en) | Video encoding method, video decoding method, video encoding device, video decoding device and storage medium | |
| US10051268B2 (en) | Method for encoding, decoding video signal and device therefor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |