KR102238091B1 - System and method for 3d model compression and decompression - Google Patents

Abstract

Embodiments of the present invention include a first input unit that receives video frame information including 3D model information encoded in a 3D model information format, and a format decoder that extracts 3D model information by analyzing the encoded 3D model information. , A compression unit for compressing 3D model information and a first output unit for converting and outputting video frame information to include compressed 3D model information, or receiving video frame information including compressed 3D model information Includes a second input unit, a decompression unit that decompresses the 3D model information, a format encoder that encodes the decompressed 3D model information into a 3D model information format, and 3D model information in which video frame information is encoded. It provides a 3D model compression and decompression system, characterized in that it includes a second output unit that converts and outputs the same.
Accordingly, embodiments of the present invention compress and decompress 3D model information irrespective of the 3D model information format, so that video frame information can be quickly transmitted, real-time playback is possible, and storage capacity can be reduced.

Description

3D model compression and decompression system and method {SYSTEM AND METHOD FOR 3D MODEL COMPRESSION AND DECOMPRESSION}

Embodiments of the present invention relate to a 3D model compression and decompression system, and more particularly, by analyzing video frame information including 3D model information of various formats, 3D model information is rapidly compressed and compressed. A 3D model compression and decompression system and method for rapidly decompressing video frame information including 3D model information and converting it into a 3D model information format.

3D video generally means reproducing an image acquired by two cameras with a stereo stereoscopic image device. However, if the 3D video is reproduced in this way, the user cannot freely change the viewing point of view.

In order to solve such a problem, a 3D video system supporting multi-view images has been developed, but as the number of views increases, there is a problem that a burden on the channel is increased.

Therefore, as a method of removing the restriction according to the viewing point of view, a method that does not use a multi-view image, but includes 3D model information for each frame of a 3D video, has emerged as a method where there is no restriction on the viewing point of the user.

Meanwhile, this type of 3D video can also be used for Augmented Reality (AR) or Virtual Reality (VR).

Here, the 3D model information generally includes a mesh constituting the shape of the surface of the 3D model, which is a polyhedron, and the mesh may mean a set of vertices and faces that define the shape of the polyhedral object.

In addition, here, the face is also called a mesh element, and a triangle is usually used, and the three points of the triangle forming the face belong to the vertices of the mesh.

However, when 3D model information is included for each frame of 3D video, the 3D model information must be stored in advance in units of frames or transmitted through a network in order to play the 3D video, so a large amount of storage is required. A problem arises in that the amount of data to be transmitted is large.

More specifically, the mesh, which is the 3D model information, is composed of the location information of the vertices (hereinafter, geometric information) and the connection information of the vertices constituting the mesh elements (hereinafter, the connection information). It increases, and the connection information that must be expressed by overlapping vertices increases more greatly. In addition, since the 3D video includes 3D model information for each frame, the total amount of increase becomes larger.

Therefore, in order to commercialize a 3D video including 3D model information for each frame, the following three conditions must be satisfied.

First, it is necessary to compress the 3D model information to reduce the storage capacity or network transmission amount of the 3D video.

Second, in order to play the compressed 3D video, the 3D model information must be decompressed for every frame of the 3D video, so the time required for decompression must be reduced to less than the time interval between 3D video frames. . For example, if you need to play 30 frames per second, you need to reduce the decompression time per frame to 33ms or less.

Third, in order to compress the 3D model information generated or updated in real time as real-time video and transmit it quickly through the network, the time required to compress the 3D model information must be reduced to less than the time interval between 3D video frames. do. For example, if 30 frames per second must be compressed and transmitted, the compression time per frame must be reduced to 33 ms or less.

Let's look at the related art.

Korean Patent Registration No. 10-1086774 relates to a low-complexity 3D mesh compression apparatus and method, which analyzes the data of the input 3D mesh model to provide vertex information, attribute information representing the characteristics of the 3D mesh model, and 3D mesh. A data analysis unit that separates connection information between vertices constituting a model, a mesh model quantization unit that generates quantized vertex information, attribute information, and connection information using vertex information, attribute information, and connection information, and quantized connection A data modulator that performs differential pulse code modulation prediction using quantization values of consecutive connection information of a three-dimensional mesh model according to the information, and data obtained by encoding quantized vertex information, attribute information, and differential pulse code-modulated connection information By including an encoding unit that outputs the 3D mesh model, the complexity of data compression for the 3D mesh model is improved, the compression rate is improved, and the compressed 3D model can be quickly and accurately restored according to the improvement of the compression complexity. Disclosed is an apparatus and method for improving a low-complexity three-dimensional mesh compression.

However, the prior art cannot provide a specific method of rapidly compressing or decompressing 3D model information in 3D video.

In order to solve the above-described problem, embodiments of the present invention provide a first input unit that receives video frame information including 3D model information encoded in a 3D model information format, and analyzes the encoded 3D model information. A 3D model compression system comprising a format decoding unit for extracting model information, a compression unit for compressing 3D model information, and a first output unit for converting and outputting the video frame information to include compressed 3D model information. Its purpose is to provide.

In addition, embodiments of the present invention are 3D model compression system, characterized in that format decoding of one frame and compression of 3D model information of another frame are performed in parallel with respect to 3D model information included in a plurality of video frame information. Its purpose is to provide.

In addition, embodiments of the present invention are 3D model information in which at least two of geometric information, connection information, attribute information, and texture information are compressed in parallel with respect to 3D model information included in one video frame information. Its purpose is to provide a model compression system.

In addition, embodiments of the present invention arrange the vertices included in the geometric information based on the order in which the mesh elements included in the connection information are arranged, or the mesh included in the connection information based on the order in which the vertices included in the geometric information are arranged. An object thereof is to provide a 3D model compression system, characterized in that elements are aligned and compressed.

In addition, embodiments of the present invention include a second input unit receiving video frame information including compressed 3D model information, a decompression unit for decompressing 3D model information, and 3D model information decompressed. An object of the present invention is to provide a 3D model decompression system comprising a format encoding unit encoding the dimensional model information format and a second output unit converting and outputting the video frame information to include the encoded 3D model information. .

In addition, embodiments of the present invention are 3D model compression, characterized in that decompression of 3D model information of one frame and format encoding of another frame are performed in parallel with respect to 3D model information included in a plurality of video frame information. Its purpose is to provide a release system.

In addition, embodiments of the present invention are characterized in that decompression of at least two information among geometric information, connection information, attribute information, and texture information is performed in parallel with respect to 3D model information included in one video frame information. Its purpose is to provide a 3D model decompression system.

In addition, embodiments of the present invention include a first input step of receiving video frame information including 3D model information encoded in a 3D model information format, and extracting 3D model information by analyzing the encoded 3D model information. A video frame including a format decoding step, a compression step of compressing 3D model information, and a first output step of converting and outputting the video frame information to include compressed 3D model information, or including compressed 3D model information The second input step of receiving information, the decompression step of decompressing the 3D model information, the format encoding step of encoding the decompressed 3D model information into the 3D model information format, and the 3rd encoding video frame information An object thereof is to provide a method for compressing and decompressing a 3D model, comprising a second output step of converting and outputting to include dimensional model information.

In addition, embodiments of the present invention may perform format decoding of one frame and compression of 3D model information of another frame in parallel for 3D model information included in a plurality of video frame information, or compress 3D model information of one frame. An object of the present invention is to provide a 3D model compression and decompression method, characterized in that decompression and format encoding of other frames are performed in parallel.

In addition, embodiments of the present invention include compression of at least two of geometric information, connection information, attribute information, and texture information in parallel for 3D model information included in one video frame information, or geometric information and connection information. An object of the present invention is to provide a 3D model compression and decompression method, characterized in that at least two pieces of information are decompressed in parallel among attribute information and texture information.

An embodiment of the present invention for achieving the above object includes: a first input unit for receiving video frame information including 3D model information encoded in a 3D model information format; A format decoder for extracting the 3D model information by analyzing the encoded 3D model information; A compression unit for compressing the 3D model information; And a first output unit converting and outputting the video frame information to include the compressed 3D model information.

In one embodiment, the 3D model compression system may perform format decoding of one frame and compression of 3D model information of another frame in parallel with respect to 3D model information included in a plurality of video frame information.

In one embodiment, the 3D model information includes at least one of geometric information, connection information, attribute information, and texture information, and the 3D model compression system is configured to perform 3D model information included in one video frame information. Compression of at least two pieces of information among geometric information, connection information, attribute information, and texture information may be performed in parallel.

In one embodiment, the 3D model compression system arranges the vertices included in the geometric information based on the order in which the mesh elements included in the connection information are arranged or the vertices included in the geometric information are arranged based on the order in which the mesh elements included in the geometric information are arranged. The mesh elements included in the connection information may be aligned and compressed.

In addition, another embodiment of the present invention for achieving the above object is a second input unit for receiving video frame information including compressed 3D model information; A decompression unit for decompressing the 3D model information; A format encoder for encoding the decompressed 3D model information into a 3D model information format; And a second output unit converting and outputting the video frame information to include the encoded 3D model information.

In an embodiment, the 3D model decompression system may perform decompression of 3D model information of one frame and format encoding of another frame in parallel with respect to 3D model information included in a plurality of video frame information.

In one embodiment, the 3D model information includes at least one of geometric information, connection information, attribute information, and texture information, and the 3D model decompression system includes 3D model information included in one video frame information. On the other hand, at least two of geometric information, connection information, attribute information, and texture information may be decompressed in parallel.

In addition, another embodiment of the present invention for achieving the above object is a first input step of receiving video frame information including 3D model information encoded in a 3D model information format; A format decoding step of extracting 3D model information by analyzing the encoded 3D model information; A compression step of compressing the 3D model information; And a first output step of converting and outputting the video frame information to include the compressed 3D model information, or a second input step of receiving video frame information including compressed 3D model information. A decompression step of decompressing the 3D model information; A format encoding step of encoding the decompressed 3D model information into a 3D model information format; And a second output step of converting and outputting the video frame information to include the encoded 3D model information.

In one embodiment, the 3D model compression and decompression method is performed in parallel with the format decoding of one frame and the compression of the 3D model information of another frame with respect to the 3D model information included in a plurality of video frame information. Decompression of the 3D model information and encoding of other frames may be performed in parallel.

In one embodiment, the 3D model information includes at least one of geometric information, connection information, attribute information, and texture information, and the 3D model compression and decompression method includes a 3D model included in one video frame information. For information, compression of at least two of the geometric information, connection information, attribute information and texture information is performed in parallel, or at least two of the geometric information, connection information, attribute information and texture information are decompressed in parallel. I can.

As described above, embodiments of the present invention extract 3D model information by analyzing the first input unit receiving video frame information including 3D model information encoded in the 3D model information format, and analyzing the encoded 3D model information. By providing a 3D model compression system, characterized in that it comprises a format decoding unit, a compression unit for compressing 3D model information, and a first output unit for converting and outputting the video frame information to include the compressed 3D model information, Regardless of the 3D model information format, video frame information can be quickly transmitted, real-time playback is possible, and storage capacity can be reduced.

In addition, embodiments of the present invention allow for fast transmission of video frame information by performing format decoding of one frame and compression of 3D model information of another frame in parallel with respect to 3D model information included in a plurality of video frame information. It can be played back in real time, and data throughput per hour can be increased.

In addition, embodiments of the present invention allow compression of at least two of geometric information, connection information, attribute information, and texture information to be performed in parallel with respect to 3D model information included in one video frame information. Can be transmitted quickly, real-time playback is possible, and data throughput per hour can be increased.

In addition, embodiments of the present invention arrange the vertices included in the geometric information based on the order in which the mesh elements included in the connection information are arranged, or the mesh included in the connection information based on the order in which the vertices included in the geometric information are arranged. By aligning and compressing elements, video frame information can be transmitted quickly, real-time playback is possible, and data throughput per hour can be increased.

In addition, embodiments of the present invention include a second input unit receiving video frame information including compressed 3D model information, a decompression unit for decompressing 3D model information, and 3D model information decompressed. By providing a 3D model decompression system, characterized in that it includes a format encoding unit encoding the dimensional model information format and a second output unit converting and outputting the video frame information to include the encoded 3D model information. Regardless of the information format, video frame information can be quickly transmitted, real-time playback is possible, and storage capacity can be reduced.

In addition, embodiments of the present invention allow the decompression of the 3D model information of one frame and the format encoding of the other frame to be performed in parallel with respect to the 3D model information included in a plurality of video frame information. It can be transmitted, can be played back in real time, and can increase data throughput per hour.

In addition, embodiments of the present invention allow the decompression of at least two information among geometric information, connection information, attribute information, and texture information to be performed in parallel with respect to 3D model information included in one video frame information. Information can be transferred quickly, real-time playback is possible, and data throughput per hour can be increased.

In addition, embodiments of the present invention include a first input step of receiving video frame information including 3D model information encoded in a 3D model information format, and extracting 3D model information by analyzing the encoded 3D model information. A video frame including a format decoding step, a compression step of compressing 3D model information, and a first output step of converting and outputting the video frame information to include compressed 3D model information, or including compressed 3D model information The second input step of receiving information, the decompression step of decompressing the 3D model information, the format encoding step of encoding the decompressed 3D model information into the 3D model information format, and the 3rd encoding video frame information By providing a 3D model compression and decompression method comprising a second output step of converting and outputting to include dimensional model information, video frame information can be quickly transmitted and reproduced in real time regardless of the 3D model information format. This is possible and the storage capacity can be reduced.

It is not limited to the above-mentioned effects, and other effects that are not mentioned can be clearly understood by those of ordinary skill in the art from the following description.

1 is a block diagram illustrating an example of a 3D model compression system according to an embodiment.
2 is a diagram illustrating an exemplary data flow in a 3D model compression system according to an exemplary embodiment.
3, 4, and 5 are diagrams illustrating an example in which format decoding and 3D model information compression are performed in a 3D model compression system according to an embodiment.
6 is a block diagram illustrating an example of a system for decompressing a 3D model according to an exemplary embodiment.
7 is a diagram illustrating an exemplary data flow in a system for decompressing a 3D model according to an exemplary embodiment.
8 is a diagram illustrating an example in which 3D model information decompression and format encoding are performed in a 3D model decompression system according to an exemplary embodiment.
9 is a diagram illustrating an example in which 3D model information decompression and format encoding are performed in a 3D model decompression system according to an exemplary embodiment.
10 is a flow chart showing a 3D model compression method according to an embodiment of the present invention.
11 is a flowchart illustrating a method of decompressing a 3D model according to an embodiment of the present invention.

Since the description of the present invention is only examples for structural or functional description, the scope of the present invention should not be construed as being limited by the embodiments described in the text. That is, since the present embodiments may have various modifications and various forms, it should be understood that the scope of the present invention includes equivalents capable of realizing the technical idea. In addition, since the object or effect presented in the present invention does not mean that a specific embodiment should include all or only such effects, the scope of the present invention should not be understood as being limited thereto.

In addition, the accompanying drawings are provided to aid the understanding of the present invention, and provide embodiments together with a detailed description. However, the technical features of the present invention are not limited to a specific drawing, and features disclosed in each drawing may be combined with each other to constitute a new embodiment.

A 3D model compression and decompression system disclosed in the following embodiments will be described in more detail with reference to each drawing.

1 is a block diagram illustrating an example of a 3D model compression system according to an embodiment.

2 is a diagram illustrating an exemplary data flow in a 3D model compression system according to an exemplary embodiment.

Referring to FIG. 1, a 3D model compression system 100 according to an embodiment includes a first input unit 110, a format decoding unit 120, a compression unit 130, a first parallel processing unit 140, and a first output. It may be configured to include the unit 150, the first transmitting and receiving unit 160, the first database 170, and the first control unit 180.

Referring to FIG. 2, the 3D model compression system 100 according to an embodiment may be used to process 3D video frames.

Here, the three-dimensional video frame does not mean a still image constituting a plurality of three-dimensional reproducible moving pictures acquired by a binocular camera or a plurality of cameras, but a series of frames including three-dimensional model information or point clouds for each frame. It may mean each frame of the configured 3D video.

The first input unit 110 receives video frame information including 3D model information encoded in the 3D model information format from another process, receives input through the first transmission/reception unit 160, or input from the first database 170 I can receive it.

Here, encoding does not mean 3D model information compression, but may mean converting 3D model information into a standard 3D model information format.

Meanwhile, the video frame information may include a time stamp or quality coding parameters for time synchronization of a video.

Here, quality coding parameters are used for adaptive bitrate streaming and may be used to adjust the quality of a stream according to a user's real-time bandwidth and CPU capacity.

In addition, the first input unit 110 may store the video frame information received as shown in FIG. 2 in an input data queue (input data buffer) provided in the first input unit 110.

The format decoder 120 may extract 3D model information by analyzing (format decoding) 3D model information included by encoding in the video frame information received by the first input unit 110 in a 3D model information format. .

Here, the 3D model information format may mean various formats such as OBJ, MTL, STL, PLY, 3DS, COLLADA, XML, VRML, X3D, FBX, U3D, O3D, etc. that express 3D model information.

In addition, here, the 3D model information may include a mesh constituting the surface shape of a polyhedral 3D model, and property information or texture information of the mesh, and the mesh is geometric information and faces (mesh) corresponding to a set of vertices. It can be configured by including connection information corresponding to a set of mesh elements).

The format decoder 120 may decompress when the format-coded 3D model information is compressed. For example, when the 3D model information is encoded in X3D format and compressed with gzip according to X3D Compressed Binary Encoding (CBD), the format decoder 120 decompresses gzip and decompresses the 3D model in X3D format. The 3D model information can be extracted by analyzing the information (format decoding).

In this way, the format decoding unit 120 may compress the 3D model information without being limited to the format of the 3D model information by format decoding the 3D model information in various formats. The compression will be described below.

The compression unit 130 may compress the 3D model information format-decoded by the format decoding unit 120. More specifically, the compression unit 130 may compress 3D model information including at least one of geometric information, connection information, attribute information, and texture information. 2 shows an example in which 3D model information includes all of geometric information, connection information, attribute information, and texture information.

Here, the geometric information and the connection information are as described above, and the attribute information may refer to a color (RGB), a normal, and the like for a vertex of a mesh or a mesh element, and the texture information may refer to a color map as a texture map.

The compression unit 130 can compress 3D model information in various ways, which will be described below.

The compression unit 130 may integrally compress the 3D model information without considering the characteristics of geometric information, connection information, attribute information, and texture information constituting the 3D model information. In this case, compression methods such as Huffman encoding and repetition length encoding may be used.

However, since this method compresses without considering characteristics such as geometric information and connection information, the compression rate may be low. Therefore, it is effective to use the 3D mesh compression method in consideration of characteristics such as geometric information and connection information, and will be described in detail below.

The geometric information may be sequentially compressed by a method such as differential pulse code modulation (DPCM), or may be compressed by dividing a space by a method such as Octree-based encoding.

The connection information may be sequentially compressed by a method such as differential pulse code modulation (DPCM), or may be compressed while rearranging the order of the connection information by a method such as Topological Surgery, Edgebreaker, and TFAN. When the order of the connection information is rearranged, the vertex information may also be rearranged and compressed.

The attribute information may be sequentially compressed by a method such as differential pulse code modulation (DPCM). However, when the order of vertices or mesh elements is changed in the process of compressing geometric information or connection information, attribute information may also need to be rearranged and compressed to match the order.

Texture information can be compressed into JPEG, PNG, AVC, or HEVC.

In addition, the compression unit 130 may determine the resolution of the 3D model information to be compressed according to quality coding parameters. That is, if the network bandwidth or CPU capacity is small, the compression unit 130 may compress a part of the geometric information and connection information related thereto by lowering the resolution of the 3D model information.

As described above, since the compression unit 130 uses an efficient compression method suitable for characteristics such as geometric information and connection information, or compresses 3D model information according to quality encoding parameters, video frame information can be quickly transmitted and real-time playback is possible. And it can reduce the storage capacity.

Meanwhile, it may take a long time for the compression unit 130 to compress 3D model information by using an efficient compression method suitable for characteristics such as geometric information and connection information.

Therefore, a method of rapidly compressing is required, which will be described in the first parallel processing unit 140 below.

The first parallel processing unit 140 may perform format decoding of one frame and compression of 3D model information of another frame in parallel with respect to 3D model information included in a plurality of video frame information. In this regard, FIGS. 3 and 4 will be described.

3 and 4 are diagrams illustrating an example in which format decoding and 3D model information compression are performed in a 3D model compression system according to an exemplary embodiment.

Referring to FIG. 3, the 3D model compression system 100 may receive video frame information every input period A (33 ms).

However, if the sum of the time B required for format decoding and the time C required for 3D model information compression is greater than the input period A of the video frame information, the output of the video frame information compressed 3D model information is the next video frame. It may be later than the input time of the information.

At this time, when the 3D model compression system 100 stores and processes newly input video frame information in an input data queue (input data buffer), if the sum of B and C is greater than A is continuously accumulated, 3D The model compression system 100 cannot process the newly input video frame information in time, so that the video frame information is continuously accumulated in the input data queue (input data buffer), and when the storage capacity of the input data queue (input data buffer) is exceeded, a new Since video frame information cannot be stored, new video frame information may be lost.

In addition, if the 3D model compression system 100 cannot process video frame information in a timely manner, real-time transmission or reproduction of 3D video may be impossible.

This problem can be solved by parallel computing in which the execution of processes is performed at the same time. Let's look at parallel computing.

Parallel computing is by connecting multiple computer systems in parallel through a communication network, or by connecting multiple processors in parallel in a single computer system, thereby dividing and assigning tasks to multiple computer systems or multiple processors in one computer system, allowing multiple processors to cooperate with each other. It can be implemented in a way that handles the task.

Here, the communication network may mean a network of a wide concept including a wired or wireless communication network.

Also here, methods of implementing parallel computing in one computer system include multi-processing and multi-threading. Multi-processing and multi-threading are as follows.

Multiprocessing creates two or more processes, and since there is no shared memory area between processes, communication between processes requires IPC (Inter Process Communication).

On the other hand, multi-threading creates two or more execution flows within a process for a single program. Each thread has a shared memory area in the thread and executes a program in cooperation to speed up the program's processing. It is a skill that increases

As described above, parallel computing includes a method of connecting multiple computer systems to a network, and a method of performing multiprocessing or multithreading in one computer system. If multiple computer systems are used, time delay may occur due to data transmission through the network. In addition, if multiprocessing is used in one computer system, there may be a time delay due to data transmission through IPC.

Therefore, it may be desirable to use a multi-threading method with a low time delay in the present invention in which a 3D video needs to be compressed quickly, and thus the multi-threading method will be focused.

However, the present invention is not limited only to the multi-threading method, and the remaining parallel computing methods may also be applied.

Referring to FIG. 4, the first parallel processing unit 140 performs format decoding in thread 0 and compresses 3D model information in thread 1, so that format decoding and transfer of one frame for a predetermined time from 33 ms, 66 ms, and 99 ms. The 3D model information compression of the frame can be performed in parallel and at the same time.

Accordingly, the first parallel processing unit 140 receives video frame information differently from FIG. 3 even when the sum of the time B required for format decoding and the time C required for compressing the 3D model information is greater than the input period A (33 ms). Can be processed in time and data throughput per hour can be increased.

The first parallel processing unit 140 may perform the following operations to enable multi-threading as shown in FIG. 4.

The first parallel processing unit 140 creates thread 0 by allocating the work of the format decoding unit 120 performing format decoding to thread 0, and performs the work of the compression unit 130 performing 3D model information compression. You can create thread 1 by assigning it to, and you can set thread 0 and thread 1 to run on different processors.

In addition, the first parallel processing unit 140 may process input video frame information while managing the state after initializing the states of threads 0 and 1 to be usable.

Specifically, when the video frame information is newly input from the first input unit 110, the first parallel processing unit 140 updates the state of thread 0 to be in use when thread 0 is available, and formats the received video frame information. Thread 0 can format-decode the 3D model information included in the code.

In addition, when the format decoding of the 3D model information is completed by thread 0, the first parallel processing unit 140 (i) updates the state of thread 0 to be usable, and (ii) thread 1 when thread 1 is in the usable state. It is possible to update the state of in use and have thread 1 compress the 3D model information with the format-decoded 3D model information.

In addition, the first parallel processing unit 140 may update the state of thread 1 to be usable when 3D model information compression is completed.

On the other hand, when the first parallel processing unit 140 uses the multi-processing method, in the above-described multi-threading method, the thread is replaced with a process, and the 3D model information decoded by the process 0 is transmitted to the process 1 through IPC. Can be implemented.

In addition, when the first parallel processing unit 140 uses a parallel computing method in which several computer systems are connected through a communication network, the 3D model information decoded by the computer 0 is replaced by a computer in the aforementioned multi-threading method. It can be implemented by having it transmitted to computer 1 through a communication network.

The first parallel processing unit 140 may perform compression of at least two information among geometric information, connection information, attribute information, and texture information for 3D model information included in one video frame information in parallel. . In this regard, FIG. 5 will be described.

5 is a diagram illustrating an example in which format decoding and 3D model information compression are performed in a 3D model compression system according to an exemplary embodiment.

5, the first parallel processing unit 140 performs format decoding in thread 0, geometric information compression and other information compression in thread 1, and connection information compression in thread 2 In addition to the compression of the 3D model information of the previous frame, the compression of geometric information, other information, and connection information of one frame can be simultaneously performed in parallel.

Here, the other information may correspond to at least one of attribute information and texture information, and the types of compression performed in threads 1 and 2 may be changed.

For example, if the geometric information compression takes the most time, the first parallel processing unit 140 may perform geometric information compression in thread 1 and connection information compression and other information compression in thread 2.

In addition, the first parallel processing unit 140 may perform geometric information compression in thread 1, connection information compression in thread 2, and other information compression in thread 3 by adding 3 to the thread.

As described above, since the first parallel processing unit 140 can reduce the time C required for 3D model information compression using the multi-threading method, the time B required for format decoding and the time C required for 3D model information compression By keeping the sum of values smaller than the input period A (33ms), the received video frame information can be processed in a timely manner and the data throughput per hour can be increased.

In addition, unlike FIG. 5, the time (C) required to compress the 3D model information for the frame input at 0 ms increases, so even if compression continues after 33 ms, the received frame at 33 ms is input at 0 ms before format decoding is completed. When the 3D model information compression for the frame is completed, the input video frame information can be processed in a timely manner.

The first parallel processing unit 140 may perform the following operations to enable multi-threading as shown in FIG. 5.

The first parallel processing unit 140 allocates the work of the format decoding unit 120 performing format decoding to thread 0 to generate thread 0, and a part of the work of the compression unit 130 performing geometric information compression and other information compression. Thread 1 can be created by allocating to thread 1, and thread 2 can be created by assigning some work of the compression unit 130 for compressing connection information to thread 2, and thread 0, thread 1, and thread 2 It can be set to run on a different processor.

In addition, the first parallel processing unit 140 may process input video frame information while managing the state after initializing the states of thread 0, thread 1, and thread 2 to be usable.

Specifically, when the video frame information is newly input from the first input unit 110, the first parallel processing unit 140 updates the state of thread 0 to be in use when thread 0 is available, and formats the received video frame information. Thread 0 can format-decode the 3D model information included in the code.

In addition, when the format decoding of the 3D model information is completed by thread 0, the first parallel processing unit 140 (i) updates the state of thread 0 to be usable and (ii) thread 1 when thread 1 is in the usable state. It is possible to update the state of in use and allow thread 1 to compress geometric information and other information on the format-decoded 3D model information.

In addition, the first parallel processing unit 140 may update the state of thread 1 to be usable when geometric information compression and other information compression are completed.

In addition, when the format decoding of the 3D model information is completed by thread 0, the first parallel processing unit 140 (i) updates the state of thread 0 to be usable and (ii) thread 2 when thread 2 is in the usable state. It is possible to update the state of in use and allow thread 2 to compress the connection information of the format-decoded 3D model information.

In addition, the first parallel processing unit 140 may update the state of the thread 2 to be usable when the compression of the connection information is completed.

In addition, the first parallel processing unit 140 may allow threads 1 and 2 to exchange data through a shared memory area.

For example, among the methods of compressing 3D model information, there is a need to change the order of the connection elements included in the connection information and at the same time change the order of the vertices included in the geometric information. In such a case, thread 2 can change the order of the vertices included in the geometric information in the shared memory area while changing the order of the connection elements included in the connection information, and thread 1 can change the order of the changed geometric information in the shared memory area. Can be compressed.

On the other hand, when the first parallel processing unit 140 uses the multi-processing method, in the above-described multi-threading method, threads are replaced with processes, and the 3D model information decoded by process 0 is transferred to process 1 or process 2 through IPC. It can be implemented by having it transmitted and having the data transmitted between Process 1 and Process 2 via IPC.

In addition, when the first parallel processing unit 140 uses a parallel computing method in which several computer systems are connected through a communication network, the 3D model information decoded by the computer 0 is replaced by a computer in the aforementioned multi-threading method. It can be implemented by having it transmitted to Computer 1 or Computer 2 through a communication network, and data between Computer 1 and Computer 2 is transmitted through a communication network.

The first parallel processing unit 140 arranges the vertices included in the geometric information based on the order in which the mesh elements included in the connection information are arranged, or the mesh included in the connection information based on the order in which the vertices included in the geometric information are arranged. Elements can be sorted. In this regard, the second parallel processing unit 240 will look at.

The first output unit 150 may convert the video frame information to include compressed 3D model information and output it through the first transmission/reception unit 160 or output to the first database 170 and store it.

In addition, the first output unit 150 may store video frame information to be output as shown in FIG. 2 in an output data queue (output data buffer) provided in the first output unit 150.

The first transmission/reception unit 160 is connected to the first input unit 110 and the first output unit 150, and the first input unit 110 receives video frame information from another terminal connected through a communication network, or a first output unit Allow 150 to transmit video frame information to other terminals.

Here, the terminal is any one of a server, a laptop computer, a mobile phone, a smart phone (2G/3G/4G/LET, smart phone), a portable media player (PMP), a personal digital assistant (PDA), and a tablet PC. I can.

The first database 170 may store video frame information to be input or may store video frame information to be output.

This database 180 is a concept including a computer-readable recording medium, and refers to not only a database of consultation, but also a database in a broad sense including data recording based on a file system, and searches for a simple set of logs. Thus, if data can be extracted, it is included within the scope of the database referred to in the present invention.

The first control unit 180 controls the overall operation of the 3D model compression system 100, and the first input unit 110, the format decoding unit 120, the compression unit 130, the first parallel processing unit 140, the first The control flow or data flow between the first output unit 150, the first transmission/reception unit 160, and the first database 170 may be controlled.

6 is a block diagram illustrating an example of a system for decompressing a 3D model according to an exemplary embodiment.

7 is a diagram illustrating an exemplary data flow in a system for decompressing a 3D model according to an exemplary embodiment.

6 and 7, the 3D model decompression system 200 according to an embodiment includes a second input unit 210, a decompression unit 220, a format encoding unit 230, and a second parallel processing unit 240. ), a second output unit 250, a second transmission/reception unit 260, a second database 270, and a second control unit 280.

Referring to FIG. 7, the 3D model decompression system 200 according to an embodiment may be used to process 3D video frames.

Here, the three-dimensional video frame does not mean a still image constituting a plurality of three-dimensional reproducible moving pictures acquired by a binocular camera or a plurality of cameras, but a series of frames including three-dimensional model information or point clouds for each frame. It may mean each frame of the configured 3D video.

The second input unit 210 may receive video frame information including compressed 3D model information through the second transmission/reception unit 260 or from the second database 270.

In addition, the second input unit 210 may store the received video frame information as shown in FIG. 7 in an input data queue (input data buffer) provided in the second input unit 210.

The decompression unit 220 may compress the 3D model information in the video frame information received by the second input unit 210 and decompress the included 3D model information.

More specifically, the decompression unit 220 may decompress 3D model information including at least one of geometric information, connection information, attribute information, and texture information. 7 shows an example in which 3D model information includes all of geometric information, connection information, attribute information, and texture information.

The decompression unit 220 can decompress the 3D model information compressed in various ways. The decompression unit 220 uses an efficient compression method suitable for characteristics such as geometric information and connection information. It may take a long time to decompress the model information.

Accordingly, a method of rapidly decompressing is required, which will be described later in the second parallel processing unit 240.

The format encoder 230 may encode the 3D model information decompressed by the decompression unit 220 into a 3D model information format.

Here, the 3D model information format may mean various formats such as OBJ, MTL, STL, PLY, 3DS, COLLADA, XML, VRML, X3D, FBX, U3D, O3D, etc. that express 3D model information.

In this way, the format encoder 230 may format-encode the 3D model information in various formats.

The second parallel processing unit 240 may perform decompression of the 3D model information of one frame and the format encoding of the other frame in parallel with respect to the 3D model information included in the plurality of video frame information. In this regard, FIGS. 8 and 9 will be described.

8 is a diagram illustrating an example in which 3D model information decompression and format encoding are performed in a 3D model decompression system according to an exemplary embodiment.

Referring to FIG. 8, the second parallel processing unit 240 performs 3D model information decompression in thread 0 and format encoding in thread 1, so that a 3D model of one frame for a predetermined time from 33 ms, 66 ms, and 99 ms. The information decompression and the format encoding of the previous frame can be performed in parallel and at the same time.

Accordingly, the second parallel processing unit 240 stores the received video frame information in a timely manner even when the sum of the time B required to decompress the 3D model information and the time C required to encode the format is greater than the input period A (33 ms). Can be processed and can increase the data throughput per hour.

The second parallel processing unit 240 may perform the following operations to enable multi-threading as shown in FIG. 8.

The second parallel processing unit 240 allocates the work of the decompression unit 220 for decompressing the 3D model information to thread 0 to generate the thread 0 and perform the work of the format encoding unit 230 for performing the format encoding. Thread 1 can be created by assigning it to thread 1, and thread 0 and thread 1 can be set to run on different processors.

In addition, the second parallel processing unit 240 may process input video frame information while managing the state after initializing the states of thread 0 and thread 1 to be usable.

Specifically, when the video frame information is newly inputted from the second input unit 210, the second parallel processing unit 240 updates the state of thread 0 to be in use when thread 0 is available, and compresses the received video frame information. Thread 0 can decompress the 3D model information.

In addition, the second parallel processing unit 240, when the 3D model information decompression is completed by thread 0, (i) updates the state of thread 0 to be usable, and (ii) when thread 1 is in the usable state, It is possible to update the state as in use and allow thread 1 to format-encode the uncompressed 3D model information.

In addition, the second parallel processing unit 240 may update the state of thread 1 to be usable when format encoding is completed.

Meanwhile, when the second parallel processing unit 240 uses the multi-processing method, in the multi-threading method described above, the thread is replaced with a process, and the 3D model information decompressed by the process 0 is transmitted to the process 1 through IPC. It can be implemented by doing.

In addition, when the second parallel processing unit 240 uses a parallel computing method in which several computer systems are connected through a communication network, 3D model information decompressed by computer 0 by replacing threads with a computer in the above-described multi-threading method. It can be implemented by having the computer 1 transmitted through the communication network.

The second parallel processing unit 240 enables decompression of at least two pieces of information from geometric information, connection information, attribute information, and texture information to 3D model information included in one video frame information in parallel. In this regard, Fig. 9 will be described.

9 is a diagram illustrating an example in which 3D model information decompression and format encoding are performed in a 3D model decompression system according to an exemplary embodiment.

9, the second parallel processing unit 240 performs geometric information decompression and other information decompression in thread 0, connection information decompression in thread 1, and format encoding in thread 2 In addition to the 3D model information decompression and the format encoding of the previous frame, the geometric information decompression and other information decompression and the connection information decompression can be simultaneously performed in parallel for one frame.

Here, the other information may correspond to at least one of attribute information and texture information, and the types of decompression performed in thread 0 and thread 1 may be changed.

For example, if it takes the most time to decompress the geometric information, the second parallel processing unit 240 can perform the decompression of the geometric information in the thread 0 and the decompression of the connection information and the other information in the thread 1. have.

In addition, the second parallel processing unit 240 may add thread 3 so that geometric information is decompressed in thread 0, connection information is decompressed in thread 1, and other information is decompressed in thread 3.

As such, the second parallel processing unit 240 can reduce the time B required to decompress the 3D model information by using the multi-threading method, and thus the time B required to decompress the 3D model information and the time B required for the format encoding. By keeping the sum of the time C smaller than the input period A (33ms), the received video frame information can be processed in a timely manner and the data throughput per hour can be increased.

However, in order to decompress the connection information for a specific mesh element in thread 1, it may be necessary to first decompress the geometric information for a specific vertex corresponding to a specific mesh element in thread 0.

For example, if you need to know the coordinates of the vertex v constituting the mesh element e in order to decompress the mesh element e included in the connection information, the v There are cases in which decompression of e must be performed first to decompress e using the coordinates of v.

In this case, if thread 1 is stopped until decompression for a specific vertex is performed in thread 0, the decompression time becomes longer.

Therefore, in order to quickly decompress without stopping the thread 1, the first parallel processing unit 140 of the 3D model compression system 100 is used to determine the geometric information based on the order in which the mesh elements included in the connection information are arranged. The included vertices may be sorted or the mesh elements included in the connection information may be sorted and compressed based on the order in which the vertices included in the geometric information are arranged.

The second parallel processing unit 240 may perform the following operations to enable multi-threading as shown in FIG. 9.

The second parallel processing unit 240 can generate thread 0 by allocating some tasks of the decompression unit 220 for decompressing geometric information and decompressing other information to thread 0, and compresses the connection information decompression. Thread 1 may be created by allocating some operations of the release unit 220 to thread 1, and thread 2 may be created by allocating the work of the format encoding unit 230 that performs format encoding to thread 2, and Threads 1 and 2 can be set to run on different processors.

In addition, the second parallel processing unit 240 may process the received video frame information while managing the state after initializing the states of thread 0, thread 1, and thread 2 to be usable.

Specifically, when the video frame information is newly inputted from the second input unit 210, the second parallel processing unit 240 updates the state of thread 0 to be in use when thread 0 is available, and compresses the received video frame information. Thread 0 can decompress geometric information and decompress other information on the included 3D model information.

In addition, the second parallel processing unit 240 may update the state of thread 0 to be usable when the geometric information decompression and other information decompression are completed.

In addition, when the video frame information is newly input from the second input unit 210, the second parallel processing unit 240 updates the state of the thread 1 to be in use when the thread 1 is in an available state, and is compressed into the received video frame information. Thread 1 may decompress the included 3D model information.

In addition, the second parallel processing unit 240 may update the state of thread 1 to be usable when decompression of the connection information is completed.

In addition, when the decompression of the 3D model information is completed by the thread 0 and the thread 1, the second parallel processing unit 240 updates the state of the thread 2 to be in use and the decompressed 3D model if the thread 2 is in an available state. You can have thread 2 format the information.

In addition, the second parallel processing unit 240 may update the state of thread 2 to be usable when format encoding is completed.

In addition, the second parallel processing unit 240 may enable thread 0 and thread 1 to exchange data through a shared memory area.

On the other hand, when the second parallel processing unit 240 uses the multi-processing method, the 3D model information decompressed by the processes 0 and 1 by replacing the thread with the process in the above-described multi-threading method is processed through IPC. It can be implemented by having the data transmitted between process 0 and process 1 via IPC.

In addition, when the second parallel processing unit 240 uses a parallel computing method in which several computer systems are connected through a communication network, the thread is replaced by a computer in the above-described multi-threading method and decompressed 3 by computer 0 and computer 1 It can be implemented by allowing the dimensional model information to be transmitted to computer 2 through a communication network and data between computer 0 and computer 1 through a communication network.

The second output unit 250 converts the video frame information to include the encoded 3D model information and outputs it to another process or outputs it to another terminal through the second transmission/reception unit 260, or the second database 270 ) To print and save.

In addition, the second output unit 250 may store video frame information to be output as shown in FIG. 7 in an output data queue (output data buffer) provided in the second output unit 250.

The second transmission/reception unit 260 is connected to the second input unit 210 and the second output unit 250, and the second input unit 210 receives video frame information from another terminal connected through a communication network or a second output unit. Allows 250 to transmit video frame information to another terminal.

The second database 270 may store video frame information to be input or may store video frame information to be output.

The second control unit 280 controls the overall operation of the 3D model decompression system 200, and the second input unit 210, the decompression unit 220, the format encoding unit 230, and the second parallel processing unit 240 , Control flow or data flow between the second output unit 250, the second transmission/reception unit 260, and the second database 270 may be controlled.

10 is a flow chart showing a 3D model compression method according to an embodiment of the present invention.

Referring to FIG. 10, a 3D model compression method 300 according to an embodiment includes a first input step (S310), a format decoding step (S320), a compression step (S330), and a first output step (S340). I can.

In the first input step (S310), the 3D model compression system 100 receives video frame information including 3D model information encoded in the 3D model information format from another process or through the first transmission/reception unit 160 It may be input or may be input from the first database 170.

In the format decoding step (S320), the 3D model compression system 100 analyzes (format decodes) the 3D model information included by encoding in the 3D model information format in the input video frame information to extract the 3D model information. I can.

Here, the 3D model information format may mean various formats such as OBJ, MTL, STL, PLY, 3DS, COLLADA, XML, VRML, X3D, FBX, U3D, O3D, etc. that express 3D model information.

In addition, the 3D model compression system 100 may decompress when the format-coded 3D model information is compressed.

In the compression step (S330), the 3D model compression system 100 may compress the format-decoded 3D model information. More specifically, the 3D model compression system 100 may compress 3D model information including at least one of geometric information, connection information, attribute information, and texture information.

Meanwhile, the 3D model compression system 100 may perform format decoding of one frame and compression of 3D model information of another frame in parallel with respect to 3D model information included in a plurality of video frame information.

That is, the format decoding step (S320) of one frame and the compression step (S330) of another frame may be simultaneously performed in parallel.

As an example for enabling such parallel computing, the 3D model compression system 100 sets thread 0 and thread 1 to be performed on different processors, allocates a format decoding operation to thread 0, and performs a 3D model information compression operation. After allocating to thread 1, you can manage the state of thread 0 and thread 1 and perform tasks in parallel.

In addition, the 3D model compression system 100 enables compression of at least two information among geometric information, connection information, attribute information, and texture information for 3D model information included in one video frame information in parallel. I can.

That is, in the compression step (S330) of one frame, geometric information compression, connection information compression, and the like may be simultaneously performed in parallel.

As an example for enabling such parallel computing, the 3D model compression system 100 sets thread 0, thread 1, and thread 2 to be performed on different processors, allocates a format decoding operation to thread 0, and compresses geometric information and After allocating the other information compression operation to thread 1 and the connection information compression operation to thread 2, it is possible to perform the operation in parallel while managing the states of thread 0, thread 1, and thread 2.

Here, the other information may correspond to at least one of attribute information and texture information.

In addition, in the compression step (S330), the 3D model compression system 100 arranges the vertices included in the geometric information based on the order in which the mesh elements included in the connection information are arranged, or the order in which the vertices included in the geometric information are arranged. The mesh elements included in the connection information can be sorted based on.

In the first output step (S340), the 3D model compression system 100 converts the video frame information to include the compressed 3D model information and outputs it through the first transmission/reception unit 160 or the first database 170 You can print it to and save it.

11 is a flowchart illustrating a method of decompressing a 3D model according to an embodiment of the present invention.

Referring to FIG. 11, a method 400 for decompressing a 3D model according to an embodiment includes a second input step (S410), a decompression step (S420), a format encoding step (S430), and a second output step (S440). Can include.

In the second input step (S410), the 3D model decompression system 200 receives video frame information including compressed 3D model information through the second transmission/reception unit 260 or input from the second database 270 I can receive it.

In the decompression step (S430), the 3D model decompression system 200 may decompress the 3D model information included by compressing the 3D model information in the received video frame information. More specifically, the 3D model decompression system 200 may decompress 3D model information including at least one of geometric information, connection information, attribute information, and texture information.

In the format encoding step S430, the 3D model decompression system 200 may encode the decompressed 3D model information into a 3D model information format.

Here, the 3D model information format may mean various formats such as OBJ, MTL, STL, PLY, 3DS, COLLADA, XML, VRML, X3D, FBX, U3D, O3D, etc. that express 3D model information.

Meanwhile, the 3D model decompression system 200 may perform parallel decompression of 3D model information of one frame and format encoding of another frame with respect to 3D model information included in a plurality of video frame information.

That is, the decompression step (S420) of one frame and the format encoding step (S430) of another frame may be simultaneously performed in parallel.

As an example for enabling such parallel computing, the 3D model decompression system 200 sets thread 0 and thread 1 to be executed in different processors, and allocates the 3D model information decompression operation to thread 0 and format-encodes it. After allocating a task to thread 1, you can manage the state of thread 0 and thread 1 while allowing them to perform tasks in parallel.

In addition, the 3D model decompression system 200 may perform parallel decompression of at least two pieces of geometric information, connection information, attribute information, and texture information for 3D model information included in one video frame information. I can do it.

That is, in the decompression step (S420) of one frame, geometric information decompression, connection information decompression, and the like may be simultaneously performed in parallel.

As an example for enabling such parallel computing, the 3D model decompression system 200 sets thread 0, thread 1, and thread 2 to be performed on different processors, and executes the geometric information decompression and other information decompression operations. After allocating to 0, decompressing the connection information to thread 1, and assigning the format encoding operation to thread 2, the state of thread 0, thread 1, and thread 2 can be managed in parallel, and the operation can be performed in parallel.

Here, the other information may correspond to at least one of attribute information and texture information.

In the second output step (S440), the 3D model decompression system 200 converts the video frame information to include the encoded 3D model information and outputs it to another process or outputs it through the second transmission/reception unit 260. Or output to the second database 270 and store.

As described above, although it has been described with reference to the preferred embodiments of the present application, those skilled in the art will variously modify the present application within the scope not departing from the spirit and scope of the present invention described in the following claims. And it will be appreciated that it can be changed.

100: 3D model compression system
110: first input unit 120: format decoding unit
130: compression unit 140: first parallel processing unit
150: first output unit
200: 3D model decompression system
210: second input unit 220: decompression unit
230: format encoding unit 240: second parallel processing unit
250: second output unit

Claims (10)

A first input unit for receiving video frame information including 3D model information encoded in a 3D model information format at each input period and storing it in an input data queue;
A format decoder for extracting the 3D model information by analyzing the encoded 3D model information;
A compression unit for compressing the 3D model information;
About 3D model information included in a plurality of video frame information
A first parallel processing unit that enables format decoding of one frame and compression of 3D model information of another frame to be performed in parallel using threads, and updates the state of each thread according to whether the format decoding and 3D model information compression are performed. ; And
A first output unit converting and outputting the video frame information to include the compressed 3D model information,
If thread 0 is in an available state, the first parallel processing unit updates thread 0 to be in use, performs format decoding using thread 0, compresses 3D model information using thread 1, and performs one frame in thread 0. 3D model compression system for updating to available when the format decoding for is finished, updating thread 1 as being in use, and compressing 3D model information for the frame in thread 1.
delete The method of claim 1,
The 3D model information includes at least one of geometric information, connection information, attribute information, and texture information,
The 3D model compression system provides 3D model information included in one video frame information.
3D model compression system, characterized in that compression of at least two pieces of information among geometric information, connection information, attribute information, and texture information is performed in parallel.
The method of claim 3,
The 3D model compression system arranges the vertices included in the geometric information based on the order in which the mesh elements included in the connection information are arranged, or includes it in the connection information based on the order in which the vertices included in the geometric information are arranged. A 3D model compression system, characterized in that the mesh elements are aligned and compressed.
A second input unit receiving video frame information including compressed 3D model information;
A decompression unit for decompressing the 3D model information;
A format encoder for encoding the decompressed 3D model information into a 3D model information format; And
About 3D model information included in a plurality of video frame information
The second that decompresses the 3D model information of one frame and performs the format encoding of the other frame in parallel using a thread, and updates the state of each thread according to whether the 3D model information is decompressed and the format encoding is performed. Parallel processing unit; And
And a second output unit converting and outputting the video frame information to include the coded 3D model information,
If thread 0 is in an available state, the second parallel processing unit updates thread 0 to be in use, decompresses 3D model information by using thread 0, performs format encoding by using thread 1, and performs one When the decompression of 3D model information for a frame is finished, it is updated to be usable, thread 1 is updated as being in use, and the 3D model decompression system enables the format encoding of the frame to be performed in thread 1.
delete The method of claim 5,
The 3D model information includes at least one of geometric information, connection information, attribute information, and texture information,
The 3D model decompression system provides 3D model information included in one video frame information.
3D model decompression system, characterized in that decompression of at least two pieces of information among geometric information, connection information, attribute information, and texture information is performed in parallel.
A first input step of receiving video frame information including 3D model information encoded in a 3D model information format;
A format decoding step of extracting 3D model information by analyzing the encoded 3D model information;
A compression step of compressing the 3D model information;
About 3D model information included in a plurality of video frame information
In the first parallel processing unit, format decoding of one frame and 3D model information compression of another frame are performed in parallel using threads, and the state of each thread is updated according to whether the format decoding and 3D model information compression are performed. Parallel compression step; And
A first output step of converting and outputting the video frame information to include the compressed 3D model information,
If thread 0 is in an available state, the first parallel processing unit updates thread 0 to be in use, performs format decoding using thread 0, compresses 3D model information using thread 1, and performs one frame in thread 0. A 3D model compression method in which, when the format decoding for is completed, it is updated to be usable, thread 1 is updated while being used, and 3D model information is compressed for the frame in thread 1.
delete The method of claim 8,
The 3D model information includes at least one of geometric information, connection information, attribute information, and texture information,
The 3D model compression and decompression method relates to 3D model information included in one video frame information.
3D model compression method, characterized in that compression of at least two pieces of information among geometric information, connection information, attribute information, and texture information is performed in parallel.

Images