KR101345544B1 - Multi-view video coding system, decoding system, bitstream extracting system for decoding base view and supporting view random access - Google Patents

Abstract

A multiview video encoding system, a decoding system, and a bitstream extraction system for supporting base view decoding and view random access are disclosed. The multiview video encoding system includes an encoder for encoding the video signal into a base view and a multiview and a base view identifier adder for adding a base view identifier for the base view to an extended parameter set for multiview video encoding. do.

Multi-view Video Coding (MVC), Advanced Video Coding (AVC), Base View, View Random Access, Encoding, Decoding

Description

MULTI-VIEW VIDEO CODING SYSTEM, DECODING SYSTEM, BITSTREAM EXTRACTING SYSTEM FOR DECODING BASE VIEW AND SUPPORTING VIEW RANDOM ACCESS}

The present invention relates to a multiview video encoding system, a decoding system and a bitstream extraction system for supporting base view decoding and view random access.

The present invention is derived from the research conducted as part of the IT growth engine technology development project of the Ministry of Information and Communication and the Ministry of Information and Telecommunications Research and Development. [Task Management Number: 2007-S-004-01] Development].

Multi-view Video Coding (MVC) is being developed to provide a more realistic three-dimensional experience to users. Multi-view Video Coding (MVC) is designed to geometrically correct and spatially process images captured by two or more cameras. It is a new field of 3D image processing technology that provides users with various viewpoint images in various directions. Such a multi-view video gives the user the opportunity to freely select a viewing point and has the advantage of being able to feel three-dimensional three-dimensional effect through a wide screen. However, multi-view video requires an efficient data processing method as the amount of data increases as the number of viewpoints increases.

Base view in defining requirements for multi-view video coding (MVC) standardization currently underway in the Moving Picture Experts Group (MPEG) and Joint Video Team (JVT). Reference is made to backward compatibility to enable decoding in the decoder of the existing Advanced Video Coding (AVC).

That is, on any arbitrary time axis, the bitstream corresponding to one time point must follow Advanced Video Coding. In other words, the multi-view video encoder can decode the video of one view from among the N views encoded by the advanced video encoding so that the multi-view video content can be shared among users with existing terminals. can do.

It also mentions view random access. In other words, multi-view video encoding should support random access in view. For example, it should be possible to access a frame at a desired point in time by minimal decoding of a frame at a different point in time. Such random point of view access is necessary to provide a smoother content to the user by allowing a short time to be accessed by an image of a random point of time in a desired time zone according to the user's request.

In order to provide backward compatibility, an embodiment of the present invention provides an image encoded with a base view in multi-view video coding (MVC) to provide an advanced video coding (AVC). Provided is a multiview video encoding system that enables decoding in a decoder.

The present invention supports an application in which a plurality of viewpoints that must be independently decoded by separately defining independent viewpoints encoded without reference from other viewpoints, and a multiview video for easily supporting view random access. Provide an encoding system.

A multiview video encoding system according to an embodiment of the present invention includes an encoder for encoding the video signal into a base view and a multiview and a base view identifier adder for adding a base view identifier for the base view to a parameter set. . Here, the base view may be limited to one view of all views.

According to an aspect of the present invention, the parameter set may be transmitted to a decoding system prior to the encoded video signal. In this case, the video signal encoded with the base view may be identified in the decoding system based on the base view identifier of the previously transmitted parameter set.

According to an aspect of the present invention, the parameter set may be included in a network abstraction layer that is transmitted before a network abstraction layer (NAL) for an actual encoded video signal when the encoded video signal is transmitted.

According to an aspect of the present invention, the encoder may further encode the video signal to an independent view, wherein the multiview video encoding system sets an independent view flag for the independent view to a network abstract layer unit (NAL unit). It may further include an independent view flag adding unit added to the header of the. Here, the video signal encoded in the independent view may be encoded without a reference from another view, and the independent view flag may indicate whether the video signal corresponding to the network abstraction layer unit is independent.

According to an aspect of the present invention, the multi-view video encoding system generates a bitstream that generates a bitstream including at least a first network abstraction layer for the parameter set and a second network abstraction layer for the encoded video signal. It may further comprise a portion, wherein the first network abstraction layer may be transmitted in advance of the second network abstraction layer.

A decoding system according to an embodiment of the present invention decodes a bitstream corresponding to a base view based on a bitstream receiver for receiving a bitstream of the encoded video signal and a parameter set included in the received bitstream. It includes a decoding unit.

A bitstream extraction system according to an embodiment of the present invention is a bitstream of at least a portion of the bitstream based on at least one of a receiver, a parameter set, and a header of a network abstract layer unit, which receives the bitstream from a multiview video encoding system. And a bitstream extracting unit for extracting the extracted bitstream and transmitting the extracted at least some bitstreams to a corresponding decoding system.

According to the present invention, in order to provide backward compatibility, an image encoded with a base view in multi-view video coding (MVC) is converted to conventional advanced video coding (AVC). Can be decoded by the decoder.

According to the present invention, it is possible to support an application in which a plurality of viewpoints to be independently decoded by separately defining independent viewpoints encoded without reference from other viewpoints, and easily support view random access.

Hereinafter, various embodiments of the present disclosure will be described in detail with reference to the accompanying drawings.

In the present specification, "Advanced Video Coding (AVC)" may mean an encoding method according to the standard of AVC / H.264. AVC / H.264 adopts Network Abstraction Layer (NAL) to improve network portability. The adoption of the NAL allows bitstreams in AVC / H.264 to be easily transmitted over many different networks. In other words, AVC / H.264 transmits this information on the network and the layer responsible for compressing the video signal so that it can be freely used in various networks, in particular, to facilitate data movement between different heterogeneous networks. It is divided into hierarchies that are responsible for converting data into specific forms. Among these, the compression layer is called a video coding layer (VCL).

The hierarchy responsible for responding is called NAL. At this time, compressed data in the form of a bitstream or a packet may be transmitted through various types of networks. The data is reconstructed in NAL units having the same data format over all networks. Table 1 below shows network abstract layer unit type codes (hereinafter, nal_unit_type) codes and their definitions in the AVC / H.264 standard. In Table 1 below, a NAL unit having a nal_unit_type of 1 to 5 is referred to as a VCL NAL unit, and the remaining NAL units are referred to as non-VCL units.

nal_unit_type

NAL unit contents and RBSP syntax structure

C

0
Unspecified

One Coded slice of a non-IDR picture
slice_layer_without _partitioning_rbsp ()
2, 3, 4
2 Coded slice data partition A
slice_data_partition _a_layer_rbsp ()
2
3 Coded slice data partition B
slice_data_partition _b_layer_rbsp ()
3
4 Coded slice data partition C
slice_data_partition _c_layer_rbsp ()
4
5 Coded slice of an IDR picture
slice_layer_without _partitioning_rbsp ()
2, 3
6 supplemental enhancement information (SEI)
sei_rbsp ()
5
7 Sequence parameter set
seq_parameter _set_rbsp ()
0
8 Picture parameter set
pic_parameter _set_rbsp ()
One
9 Access unit delimiter
access_unit_delimiter _rbsp ()
6
10 End of sequence
end_of_seq_rbsp ()
7
11 End of stream
end_of_stream_rbsp ()
8
12 Filter data
filter_data_rbsp ()
9
3 ... 23
Reserved
23 ... 31
Unspecified

Content using Multi-view Video Coding (MVC) can be a decoder of Advanced Video Coding in which the encoded bitstream decodes only video of one view or a decoder of Multiview Video Coding in decoding all of multiview video. Can be entered. In this case, the form of the bitstream may be classified into two cases as follows.

1. When a bitstream encoded by multi-view video encoding is input to the decoder as a whole bitstream (no bitstream extractor).

2. The bitstream encoded by multi-view video encoding is input to the decoder as a bitstream in which only the required NAL is extracted using information on the NAL unit (if there is a bitstream extractor).

1 is an example of an application system in which a bitstream encoded by multiview video encoding is input to a decoder as a whole. That is, FIG. 1 is an example showing the first case of the two cases described above, and is located at the rear end of the multiview video encoding encoder 101 or the front end of the multiview video encoding decoder 102 or the advanced video encoding decoder 103. The case where there is no bitstream extractor. In this case, the multi-view video encoding decoder 102 may decode the input bitstream without any problem.

However, the advanced video encoding decoder 103 is unable to decode at one time when the number of base views in the encoded bitstream is input. This is because a prefix NAL unit is preceded by a coded slices unit for the base view. 2 is an example of an encoded slice unit. That is, in multi-view video encoding, the coded slices unit of the base view set to 'nal_unit_type = 1 to 5' 201 is of the same type as 'nal_unit_type = 14' (202) as shown in FIG. 2. The prefix NAL unit is preceded. Here, 'nal_unit_type = 14' 202 may be defined to indicate an encoded slice unit for a base view in multi-view video encoding through nal_unit_type reserved among nal_unit_type which is an identifier indicating a type of NAL unit, and may be defined as' nal_unit_header_svc_mvc_extention. Only () 'can be set.

In other words, when the prefix NAL unit is defined in the multi-view video encoding encoder 101 as shown in the example of FIG. 2, the 'nal_unit_header_svc_mvc_extension ()' syntax, which defines header information required for multi-view video decoding, is executed. Information of a view identifier 'view_id' is obtained from the encoded slice unit of the base view following the NAL unit. However, since the advanced video encoding decoder 103 does not recognize the prefix NAL unit as shown in row 15 of Table 1, the advanced video encoding decoder 103 does not obtain the viewpoint identifier for a subsequent NAL unit, and simply 'nal_unit_type = 1 to 5'. The encoded slice unit set to '201 is decoded. In this case, when the number of the base views is plural as described above, the encoded slice units set to 'nal_unit_type = 1 to 5' 201 are advanced video because bitstreams of two or more views are encoded with reference to each other. The encoding decoder 103 cannot normally decode the bitstream. Therefore, in multi-view video encoding, only one base view should be set. However, depending on the application of multi-view video encoding, there are applications in which a plurality of viewpoints to be independently decoded exist.

FIG. 3 is an example of an application system in which a bitstream encoded by multi-view video encoding is input to a decoder as a bitstream from which only necessary NALs are extracted using information on a NAL unit. That is, FIG. 3 is an example showing the second case of the two cases described above, and is located at the rear end of the multiview video encoding encoder 101 or the front end of the multiview video encoding decoder 102 or the advanced video encoding decoder 103. It shows the case where the bitstream extractor is present. In this case, as described above, in the bitstream encoded by the multiview video encoding encoder 301, a bitstream for a required NAL is extracted by using the information on the NAL unit in the bitstream extractor 302 to decode the multiview video encoding. Is input to the device 303 or the advanced video encoding decoder 304.

Even in this case, when the bitstream output from the multiview video encoding encoder 301 is input to the multiview video encoding decoder 303, the bitstream may be decoded without any problem, and the advanced video encoding decoder 304 may be used. In order to support an application in which a plurality of time points that need to be independently decoded are supported, the bitstream extractor 302 extracts a bitstream of a video signal encoded to the independent time point through the information on the independent time point. It must be supported to be input to the video encoding decoder 304.

Therefore, in an embodiment of the present invention, the base view identifier for the base view is added to the extended sequence parameter set for the multiview video encoding to support backward compatibility with the advanced video encoding. In addition, similarly to the base view, an independent view that is encoded without a reference from another view is separately defined to support an application that requires a plurality of views to be decoded, and to facilitate access of a view randomly.

4 is a block diagram illustrating an internal configuration of a multiview video encoding system according to an embodiment of the present invention. Here, the multi-view video encoding system 400 according to an embodiment of the present invention encodes video signals photographed through a plurality of cameras through multi-view video encoding. In this case, the multi-view video encoding system 400 includes an encoder 401, a base view identifier adder 402, and an independent view flag adder 403, as shown in FIG. 4.

The encoder 401 encodes the video signal into a base view and a multiview. In this case, the base view may be limited to one view of all views, and a video signal corresponding to the base view may be encoded without reference from another view.

The base view identifier adder 402 adds a base view identifier for the base view to the parameter set. In this case, the parameter set may be transmitted to the decoding system 410 prior to the encoded video signal. That is, the video signal encoded with the base view can be identified in the decoding system based on the base view identifier of the previously transmitted parameter set. In more detail, the parameter set may be included in the NAL transmitted before the NAL for the actual encoded video signal when the encoded video signal is transmitted.

5 is an example of a bitstream structure. In FIG. 5, the bitstream 500 includes five NALs, and each NAL includes a NAL unit and a raw byte sequence payload (RBSP). Here, the RBSP may include a slice set corresponding to a parameter set or a VCL indicating information such as a sequence parameter set (SPS) and a picture parameter set (PPS). In addition, the header of the NAL unit may include a view identifier of a corresponding video signal.

In other words, the base view identifier adder 402 adds the base view identifier for the base view to the same parameter set as the SPS transmitted before the NAL for the actual encoded video signal, thereby receiving the bitstream. After 410 obtains the base view identifier, it may compare the view identifier of the header of the NAL unit to efficiently check the NAL for the video signal encoded at the base view.

Table 2 shows the 'seq_parameter_set_mvc_extension ()' syntax of the Joint Multi-view Video Model (JMVM) as defined by the Moving Picture Experts Group (MPEG) and the Joint Video Team (JVT). The syntax defines a base view identifier (base_view_id) for a base view limited to one view (line 3 in Table 2). That is, as shown in Table 2, the conventional advanced video encoder can decode the video signal encoded at the base view through the base view identifier without parsing. That is, it is possible to efficiently provide backward compatibility between multiview video encoding and advanced video encoding.

seq_parameter_set_mvc_extension () { C Descriptor mum _ views _ minus _One ue (v) base _ view _ id ue (v)   for (i = 0; i <= num_views_minus_1; i ++) view _ id [i] ue (v)   for (i = 0; i <= num_views_minus_1; i ++) { num _ anchor _ refs _ l0 [i] ue (v)     for (j = 0; j <= num_anchor_refs_l0 [i]; j ++) anchor _ ref _ l0 [i] [j] ue (v) num _ anchor _ refs _ l1 [i] ue (v)     for (j = 0; j <= num_anchor_refs_l1 [i]; j ++) anchor _ ref _ l0 [i] [j] ue (v)   }   for (i = 0; i <= num_views_minus_1; i ++) { num _ non _ anchor _ refs _ l0 [i] ue (v)     for (j = 0; j <= num_ non_anchor_refs_l0 [i]; j ++) non _ anchor _ ref _ l0 [i] [j] ue (v) num _ non _ anchor _ refs _ l1 [i] ue (v)     for (j = 0; j <= num_ non_anchor_refs_l1 [i]; j ++) non _ anchor _ ref _ l1 [i] [j] ue (v)   } }

Here, 'num_views_minus_1' may mean a number smaller than the total number of views encoded in the bitstream. For example, the 'num_views_minus_1' may mean a number in a range of 0 to 1023.

In addition, the 'base_view_id' may mean 'view_id' of the base view, and the 'view_id [i]' may represent 'view_id' of a view having an encoding order indicated by the variable 'i'.

'Num_anchor_refs_' l0 [i] 'may represent the number of possible inner-view prediction references for' RefPicList () ', and' RefPicList () 'is the anchor picture having the same' view_id 'as the' view_id [i] '. It can mean a syntax used for anchor ficture). At this time, the 'num_anchor_refs_ The value of l0 [i] 'may be equal to or less than'num_ref_frames', which is the number of reference frames.

'Ue (v)' may mean an unsigned integer Exp-Golomb code with v-bits.

In addition, the encoder 401 may further encode the video signal to an independent view. In this case, the independent view flag adder 403 illustrated in FIG. 4 may set an independent view flag for the independent view as a header of a NAL unit. Add to In this case, the video signal encoded in the independent view may be encoded without reference from another view. This is because there are applications having a plurality of time points that need to be independently decoded, and to support view point random access for the applications.

Table 3 shows the syntax of 'nal_unit_header_svc_mvc_extension ()' among the syntaxes of the federated multi-view video model defined by the moving image group and the federation video team. Indicates. That is, Table 4 shows an example in which the independent view flag is added to header information in the multi-view video encoding system 400 according to an embodiment of the present invention.

nal_unit_header_svc_mvc_extension () { C Descriptor svc _ mvc _ flag ALL u (1)   if (! svc_mvc_flag) { idr _ flag ALL u (1) priority _ id ALL u (6) temporal _ id ALL u (3) dependency _ id ALL u (3) quality _ id ALL u (2) layer _ base _ flag ALL u (1) use _ base _ prediction _ flag ALL u (1) discardable _ flag ALL u (1) output _ flag ALL u (1) reserved _ zero _ four _ bits ALL u (3)   } else { priority _ id ALL u (6) temporal _ id ALL u (3) anchor _ pic _ flag ALL u (1) view _ id ALL u (10) idr _ flag ALL u (1) inter _ view _ flag ALL u (1) reserved _ zero _ one _ bit ALL u (1)   }   nalUnitHeaderBytes + = 3 }

nal_unit_header_svc_mvc_extension () { C Descriptor svc _ mvc _ flag ALL u (1)   if (! svc_mvc_flag) { idr _ flag ALL u (1) priority _ id ALL u (6) temporal _ id ALL u (3) dependency _ id ALL u (3) quality _ id ALL u (2) layer _ base _ flag ALL u (1) use _ base _ prediction _ flag ALL u (1) discardable _ flag ALL u (1) output _ flag ALL u (1) reserved _ zero _ four _ bits ALL u (3)     nalUnitHeaderBytes + = 3   } else { priority _ id ALL u (6) temporal _ id ALL u (3) inter _ view _ flag ALL u (1) If (nal_unit_type == 20) { anchor _ pic _ flag ALL u (1) view _ id ALL u (10) idr _ flag ALL u (1) Independent _ view _ flag ALL u (1) nalUnitHeaderBytes + = 3 } else { reserved _ zero _ five _ bits ALL u (5) nalUnitHeaderBytes + = 2 }   } }

Here, the 'priority_id' may indicate a priority for the NAL unit. The smaller 'priority_id' may mean higher priority. If the NAL unit is a prefix NAL unit, the priority identifier may be applied to the combined NAL unit immediately following the prefix NAL unit in decoding order. When any possible value of the 'priority_id' is called 'pid', all VCL NAL units having a 'priority_id' greater than the 'pid' as well as a non-VCL (Video Coding Layer) NAL unit and a Supplemental Enhancement Information (SEI) message The bitstream obtainable by discarding may follow the recommended international standard.

The 'temporal_id' may mean a level of temporary scalability of the NAL unit. If there is no 'temporal_id' for one NAL, the value of 'temporal_id' may be estimated as the value of 'temporal_id' of the combined prefix network layer step.

When the 'inter_view_flag' has a value of 0, the picture encoded in the current NAL unit is not used for internal view prediction. When the 'inter_view_flag' has a value of 1, the encoded picture is used for the internal view prediction. Can mean.

The 'reserved_zero_one_bit' and the 'reserved_zero_five_bit' will have a value of zero. The other values of the 'reserved_zero_one_bit' and the 'reserved_zero_five_bit' are later described in ITU-T | Can be defined by ISO / IEC. The decoder may ignore the values of 'reserved_zero_one_bit' and 'reserved_zero_five_bit'.

When the basic time point is limited to one time point, 'anchor_pic_flag (1bit)', 'view_id (10bit)', and 'idr_flag (1bit)', which are previously defined in the syntax 'nal_unit_header_svc_mvc_extension ()', as shown in Table 3, are NAL. Since the value can be known by the 'nal_unit_type' value indicating the type of the unit, the coded bit can be reduced by changing as shown in Table 4.

That is, information about the 'anchor_pic_flag (1bit)' may be obtained by 'slice_type' indicating a slice type in the NAL set to 'nal_unit_type = 1 to 5' along the prefix NAL unit set to 'nal_unit_type = 14'. For example, when the values of 'slice_type' are 2, 4, 7, and 9, the value of 'anchor_pic_flag' has a value of 1, and if not, the value of 'anchor_pic_flag' has a value of 0.

In addition, if the 'seq_parameter_set_mvc_extension ()' syntax includes information for the basic time identifier of the basic time, it is easy to know that the corresponding NAL corresponds to the basic time through the header of the NAL unit set to 'nal_unit_type = 1 to 5'. Can be. Thus, the prefix NAL unit does not need information for the base view identifier.

In addition, the 'idr_flag (1 bit)' can be known by the 'nal_unit_type' following the 'nal_unit_type = 14'. That is, when the 'nal_unit_type = 5', the slice is an Instantaneous Decoding Refresh (IDR) picture, otherwise, the slice is a non-IDR picture.

The independent NAL unit encoded independently without inter-view prediction may be set to 'nal_unit_type = 20', which is one of the reserved nal_unit_type, in which case the independent point is checked by checking only the header of the NAL unit. You can check whether or not. The NAL unit encoded at the base view may be set as 'nal_unit_type = 1 to 5'. Therefore, the NAL unit header may include an independent view flag for checking whether the viewpoint is an independent view as shown in Table 3. Here, the independent view flag may be expressed as 'independent_view_flag' as shown in Table 4. That is, the independent stream flag enables the bitstream extraction system according to an embodiment of the present invention to easily manage and process a specific independent view point without parsing the entire view dependency information from header information such as 'nal_unit_header_svc_mvc_extension ()' syntax. Will be.

In addition, the independent time flag is useful for finding independent time when the type of NAL unit is 'nal_unit_type = 20'. If the type of the NAL unit is' nal_unit_type = 20 'in a decoding system such as an advanced video encoding decoder, and is to be decoded by selecting one time point that is an independent view, the independent view flag, the' By resetting 'nal_unit_type = 20' to 'nal_unit_type = 1 to 5' based on the idr_flag 'and the view identifier, the view can be decoded in the same way as the base view in a decoding system such as an advanced video encoding decoder.

'U (n)' may mean an unsigned integer having n-bits.

Table 5 shows the existing 'nal_unit ()' syntax among the syntaxes of the combined multi-view video model defined by the moving picture expert group and the combined video team, and Table 6 changes the existing code according to the reduced coding bit as shown in Table 4 above. Represents a syntax change from the 'nal_unit ()' syntax.

nal_unit (NumBytesInNALunit) { C Descriptor forbidden _ zero _ bit ALL f (1) nal _ ref _ idc ALL u (2) nal _ unit _ type ALL u (5)   nalUnitHeaderBytes = 1   if (nal_unit_type == 14 nal_unit_type == 20) {     nal_unit_header_svc_mvc_extension ()     nalUnitHeaderBytes + = 3   }   NumBytesInRBSP = 0   for (I = nalUnitHeaderBytes; i <NumBytesInNALunit; i ++) {     if (i + 2 <NumBytesInNALunit && next_bits (24) == 0x000003) { rbsp _ byte [ NumBytesInRBSP ++] ALL b (8) rbsp _ byte [ NumBytesInRBSP ++] ALL b (8)       i + = 2       emulation_prevention_three_byte / * equal to 0x03 * / ALL f (8)     } else rbsp _ byte [ NumBytesInRBSP ++] ALL b (8)   } }

nal_unit (NumBytesInNALunit) { C Descriptor forbidden _ zero _ bit ALL f (1) nal _ ref _ idc ALL u (2) nal _ unit _ type ALL u (5)   nalUnitHeaderBytes = 1   if (nal_unit_type == 14 nal_unit_type == 20) {     nal_unit_header_svc_mvc_extension () if (nal_unit_type == 14) nalUnitHeaderBytes + = 2 else nalUnitHeaderBytes + = 3   }   NumBytesInRBSP = 0   for (I = nalUnitHeaderBytes; i <NumBytesInNALunit; i ++) {     if (i + 2 <NumBytesInNALunit && next_bits (24) == 0x000003) { rbsp _ byte [ NumBytesInRBSP ++] ALL b (8) rbsp _ byte [ NumBytesInRBSP ++] ALL b (8)       i + = 2 emulation _ prevention _ three _ byte / * equal to 0x03 * / ALL f (8)     } else rbsp _ byte [ NumBytesInRBSP ++] ALL b (8)   } }

Here, 'f (n)' may mean data of a fixed pattern having n-bits, and 'b (8)' may mean data of a byte type having 8-bits.

The multi-view video encoding system 400 may further include a bitstream generator (not shown) for generating a bitstream including at least a first NAL for the parameter set and a second NAL for the encoded video signal. have. The bitstream generated as described above may be directly transmitted to another decoding system 410 according to an embodiment of the present invention, or may be transmitted to the decoding system 410 through a bitstream extraction system (not shown). In this case, the first NAL may be transmitted in advance of the second NAL.

6 is a block diagram illustrating an internal configuration of a decoding system according to an embodiment of the present invention. As illustrated in FIG. 6, the decoding system 600 according to an embodiment of the present invention may include a bitstream receiver 601 and a decoder 602.

The bitstream receiver 601 receives a bitstream of the encoded signal. In this case, the received bitstream is one bitstream of the bitstream from which only the required NAL is extracted by using the entire bitstream for the encoded signal initially transmitted from the multiview video encoding system 610 and the information on the NAL unit. It may include. In addition, the received bitstream may also include a NAL including a NAL unit.

The decoder 602 decodes the bitstream corresponding to the base view based on the parameter set included in the received bitstream. In this case, the bitstream for the parameter set may be received prior to the bitstream for the encoded video signal. In addition, the parameter set may include a base view identifier for the video signal encoded as the base view, wherein the decoder 602 uses the base view identifier to perform the encoding of the encoded video signal in the bitstream. The video signal encoded at the base view may be identified and decoded. That is, it is possible to easily find and decode the bitstream of the video signal encoded at the base view through the base view identifier included in the parameter set.

In addition, the decoder 602 may selectively identify and decode a video signal encoded at an independent view among the encoded video signals in the bitstream based on the independent view flag included in the header of the NAL unit. That is, since only one basic view is set, but an application that requires a plurality of independently decoded views is required, the independent view that is encoded without a reference from another view is checked through the independent view flag. It can also support point-in-time random access.

Therefore, even if the decoding system 600 uses advanced video encoding, the encoded content can be decoded through multi-view video encoding. In addition, since the number of viewpoints during encoding is not limited, the independent viewpoint can support an application that requires a plurality of viewpoints to be independently decoded, and an arbitrary viewpoint random access is also easy.

7 is a block diagram illustrating an internal configuration of a bitstream extraction system according to an embodiment of the present invention. As shown in FIG. 7, the non-stream stream extracting system 700 according to an embodiment of the present invention includes a receiver 701, a bitstream extractor 702, and a transmitter 703.

The receiver 701 receives the encoded bitstream from the multi-view video encoding system 710. In this case, the encoded bitstream includes at least one NAL.

The bitstream extractor 702 extracts a bitstream of at least a portion of the bitstream based on at least one of a parameter set and a header of the NAL unit. Here, the parameter set may be included in the NAL transmitted before the NAL for the actual encoded video signal. Also, the parameter set may include a base view identifier for a video signal encoded with a base view, and in this case, the bitstream extractor 702 may encode a video signal encoded with the base view based on the base view identifier. It is possible to extract a bitstream for. In addition, the header may include an independent view flag for the video signal encoded at the independent view, and the independent view flag may indicate whether the encoded video signal corresponding to the NAL unit is independent. In this case, the bitstream extractor 702 may extract a bitstream of the video signal encoded at the independent view based on the independent view flag.

The transmitter 703 transmits the extracted at least some bitstreams to the corresponding decoding system 720. That is, even when the decoding system 720 receives the advanced video decoding system, the decoding system 720 normally receives the bitstream for the independently encoded base view or the independently encoded independent view. Or support for an application having a plurality of time points that need to be independently decoded and support for a random access time point.

In addition, the bitstream extraction system 700 may further include a type change unit (not shown) for changing the type of the NAL unit for the video signal encoded at the independent view. That is, by selectively changing the type of the NAL unit for the video signal encoded in the independent view to the type of the NAL unit for the video signal encoded in the base view, the video signal encoded in the independent view in the decoding system 720 May be decoded in the same manner as the video signal encoded at the base view.

As described above, the present invention has been described by specific embodiments such as specific components and the like. For those skilled in the art, various modifications and variations are possible from these descriptions.

Accordingly, the spirit of the present invention should not be construed as being limited to the embodiments described, and all of the equivalents or equivalents of the claims, as well as the following claims, belong to the scope of the present invention .

1 is an example of an application system in which a bitstream encoded by multiview video encoding is input to a decoder as a whole.

2 is an example of an encoded slice unit.

FIG. 3 is an example of an application system in which a bitstream encoded by multi-view video encoding is input to a decoder as a bitstream from which only necessary NALs are extracted using information on a NAL unit.

4 is a block diagram illustrating an internal configuration of a multiview video encoding system according to an embodiment of the present invention.

5 is an example of a bitstream structure.

6 is a block diagram illustrating an internal configuration of a decoding system according to an embodiment of the present invention.

7 is a block diagram illustrating an internal configuration of a bitstream extraction system according to an embodiment of the present invention.

<Explanation of symbols for the main parts of the drawings>

400: Multiview Video Coding System

401: encoder

402: adding a base view identifier

403: Independence view flag addition unit

Claims (24)

In a multiview video encoding system for encoding a video signal photographed through a plurality of cameras through a multiview video encoding, An encoder which encodes the video signal into a base view and a multiview; And A base view identifier adding unit for adding a base view identifier for identifying a video signal of the base view to a parameter set, The video signal of the basic view is a video signal that is encoded without using the video signal of another viewpoint. The method of claim 1, The parameter set is transmitted to a decoding system prior to the encoded video signal. 3. The method of claim 2, And the video signal encoded with the base view is identified in the decoding system based on a base view identifier of a previously transmitted parameter set. The method of claim 1, And the parameter set is included in a network abstraction layer transmitted in advance of a network abstraction layer (NAL) for an actual encoded video signal upon transmission of the encoded video signal. 5. The method of claim 4, The parameter set includes a sequence parameter set (SPS), And the sequence parameter set is a network abstraction layer that is first transmitted upon bitstream transmission for the encoded video signal. delete delete delete The method of claim 1, The base view is a multi-view video encoding system, characterized in that it is limited to one of the viewpoints. The method of claim 1, A bitstream generator for generating a bitstream including at least a first network abstraction layer for the parameter set and a second network abstraction layer for the encoded video signal More, And wherein the first network abstraction layer is transmitted prior to the second network abstraction layer. A decoding system for decoding a video signal encoded through multiview video encoding, A bitstream receiver configured to receive a bitstream of the encoded video signal; And A decoder configured to decode a video signal encoded at a base view based on a parameter set included in the received bitstream, The decoder may identify a video signal encoded with the base view in the bitstream by using a base view identifier included in the parameter set, And the video signal encoded at the base view is a video signal encoded without using video signals at different views. 12. The method of claim 11, And the bitstream for the parameter set is received prior to the bitstream for the encoded video signal. delete delete 12. The method of claim 11, And the received bitstream comprises one bitstream of the entire bitstream for the encoded signal and a bitstream from which only the necessary network abstraction layer is extracted using information about a network abstraction layer unit. A receiver for receiving a bitstream from a multiview video encoding system; A bitstream extracting unit for extracting a bitstream of at least a portion of the bitstream based on at least one of a parameter set and a header of a network abstract layer unit; And And a transmitter for transmitting the extracted at least some bitstreams to a corresponding decoding system. The parameter set includes a base view identifier for identifying a video signal encoded with a base view, And the video signal encoded at the base view is a video signal encoded without using a video signal of another view. 17. The method of claim 16, The encoded bitstream includes at least one network abstraction layer, And the parameter set is included in a network abstraction layer that is transmitted prior to the network abstraction layer for the actual coded video signal. 17. The method of claim 16, The bit stream extractor, And extracting a bitstream for the video signal encoded with the base view based on the base view identifier. 17. The method of claim 16, The header includes an independent view flag for a video signal encoded with independent view, And the independent view flag indicates whether the encoded video signal corresponding to the network abstract layer unit is independent. 20. The method of claim 19, The bit stream extractor, And extracting a bitstream for a video signal encoded with independent time based on the independent time flag. 20. The method of claim 19, A type change unit for changing a type of a network abstraction layer unit for the video signal encoded in the independent view Bitstream extraction system further comprising. In a multiview video encoding system for encoding a video signal photographed through a plurality of cameras through a multiview video encoding, An encoder which encodes the video signal into a multiview and an independent view; An independent view flag adding unit for adding an independent view flag for identifying the independent video signal to a header of a network abstract layer unit (NAL unit), The independent video signal is a multi-view video encoding system, which is a video signal encoded without using a video signal of another viewpoint. 23. The method of claim 22, The independent view flag indicates whether or not the video signal corresponding to the network abstraction layer unit is independent. A decoding system for decoding a video signal encoded through multiview video encoding, A bitstream receiver configured to receive a bitstream of the encoded video signal; And A decoder configured to selectively decode a video signal encoded at an independent view included in the received bitstream, The decoder may identify a video signal encoded with the independent view in the bitstream based on an independent view flag included in a header of a network abstract layer unit, The video signal encoded at the independent time point is a video signal coded without using video signals at different time points.

Images