对比文件名称：2013-09-12_US2013235152A_发明申请_US20130235152A1 Video Coding and Decoding
目标专利名称：237在多层译码中译码恢复点补充增强信息(SEI)消息和区刷新信息SEI消息的方法CN105850138B
本次调用的模型名称：gpt-4o
### 特征比对表格
| 技术特征描述以及公开性判断结果 | 对比文件原文引用 | 公开性论述 |
| **技术特征A**：使恢复点补充增强信息SEI消息与包含在存取单元内的多个图片当中的一个图片相关联。<br>**判断结果**：隐含公开(<<<a>>>) | [0320] “The encoder may also generate a recovery point SEI message encapsulated into the MVC scalable nesting SEI message. The MVC scalable nesting SEI message indicates the view_id values for which the contained SEI message applies to. When a recovery point SEI message is contained in an MVC scalable nesting SEI message, the semantics of the contained recovery point SEI message may be interpreted in such a manner that decoding of only the views indicated in the MVC scalable nesting SEI message are required to recovery the indicated views, i.e. no inter-view or view synthesis prediction takes place from other views than those indicated in the MVC scalable nesting SEI message. Hence, view-wise entry point (a.k.a. random access point) and recovery point information can be conveyed when a recovery point SEI message is encapsulated into a MVC scalable nesting SEI message.”<br>[0321] “The encoder may generate an MVC recovery point SEI message providing view-wise information of entry points (a.k.a. random access point) and recovery points.” | 对比文件公开了在多视图编码（MVC）中，恢复点SEI消息可以被封装在MVC可缩放嵌套SEI消息中，该嵌套消息指明了该恢复点SEI消息所适用的视图（view_id）。这实质上建立了恢复点SEI消息与特定视图（对应于多层位流的一个层）的关联。根据目标专利说明书[0146]-[0147]段，多层位流（如多视图或可缩放视频）的一个存取单元包含多个图片（每个层一个图片）。因此，使恢复点SEI消息通过视图ID与一个特定视图关联，本领域技术人员可以毫无疑义地推理出，这等价于使该消息与该视图在存取单元内的对应图片相关联。对比文件虽然没有使用“图片”一词，但视图组件（view component）即是对应视图在某个存取单元的图片（参见[0150]段）。因此，该技术特征被对比文件隐含公开。 |
| **技术特征B**：所述存取单元包含在所述多层位流内。<br>**判断结果**：直接公开(<<<B>>>) | [0143] “An access unit in MVC is defined to be a set of NAL units that are consecutive in decoding order and contain exactly one primary coded picture consisting of one or more view components. ... In other words, an access unit in MVC may contain the view components of the views for one output time instance.”<br>[0122] “In H.264/AVC, an access unit includes a primary coded picture and those NAL units that are associated with it.” | 对比文件明确定义了多视图视频编码（MVC）中的存取单元概念，其包含一个或多个视图组件（即多个图片），这些视图组件属于多层位流（多视图位流）。该描述与目标专利中“存取单元包含在多层位流内”的特征完全一致。因此，该技术特征被对比文件直接公开。 |
| **技术特征C**：所述相关联的恢复点SEI消息不包含在可缩放嵌套SEI消息中。<br>**判断结果**：隐含公开(<<<c>>>) | [0320] “The encoder may also generate a recovery point SEI message encapsulated into the MVC scalable nesting SEI message...”<br>[0319] “The encoder may generate an indication of a VRA access unit into the bitstream through various syntax elements including but not limited to the following. The encoder may generate a flag or other syntax element in the NAL unit header or in the slice header. ... The encoder may also generate a recovery point SEI message encapsulated into the MVC scalable nesting SEI message.” | 对比文件明确教导了恢复点SEI消息“可以”被封装在MVC可缩放嵌套SEI消息中（一种具体实施方式）。然而，对比文件并未限定恢复点SEI消息必须被如此封装。本领域技术人员知晓，SEI消息作为语法元素可以独立存在于位流中，而不必被嵌套。根据对比文件[0319]段，指示VRA访问单元的方式有多种，包括在NAL单元头或片头中使用标志或其他语法元素。因此，本领域技术人员可以合理地推断，恢复点SEI消息本身也可以不通过嵌套SEI消息的方式直接与特定视图/图片相关联。虽然对比文件没有明确记载“不包含在可缩放嵌套SEI消息中”这一否定性特征，但根据其公开的多种指示方式，该技术方案是可能被本领域技术人员推理得出的。因此，该技术特征被对比文件隐含公开。 |
| **技术特征D**：可缩放嵌套SEI消息是恢复点SEI消息的容器且指定了所含有的恢复点SEI消息应用于的多个层。<br>**判断结果**：直接公开(<<<D>>>) | [0156] “MVC scalable nesting SEI message may be used to contain other SEI messages, which were typically specified for single-view (H.264/AVC) use. The MVC Scalable Nesting SEI message indicates the view_id values for which the contained SEI messages apply to.”<br>[0320] “The encoder may also generate a recovery point SEI message encapsulated into the MVC scalable nesting SEI message. The MVC scalable nesting SEI message indicates the view_id values for which the contained SEI message applies to...” | 对比文件明确描述了MVC可缩放嵌套SEI消息可以作为容器，包含其他SEI消息（如恢复点SEI消息），并且指明了所包含的SEI消息所适用的视图ID（view_id）。视图（view）即对应于多层位流中的层（layer），如目标专利说明书[0025]段所述。因此，“指定了所含有的恢复点SEI消息应用于的多个层”这一特征被对比文件直接公开。 |
| **技术特征E**：所述相关联的恢复点SEI消息指示在所述一个图片之前或所述一个图片处的随机存取点处开始解码之后作为恢复点的随后图片。<br>**判断结果**：直接公开(<<<E>>>) | [0256] “A random access point or an entry point is here defined as a point in the bitstream where the decoding process can be initiated. A recovery point is defined as a point in the bitstream at which the recovery of an exact or an approximate representation of the decoded pictures represented by the bitstream may be achieved after a random access or broken link at the preceding random access point.”<br>[0260] “An H.264/AVC decoder can recognize an intra picture starting an open GOP from the recovery point SEI message in the H.264/AVC bitstream. When a recovery point SEI message is used, the random access point is the access unit containing the recovery point SEI message. The recovery point (of output pictures in output order) is indicated by the recovery_frame_cnt syntax element of the recovery point SEI message.”<br>[0320] “Hence, view-wise entry point (a.k.a. random access point) and recovery point information can be conveyed when a recovery point SEI message is encapsulated into a MVC scalable nesting SEI message.” | 对比文件详细描述了恢复点SEI消息的功能：它存在于一个存取单元（随机存取点）中，并指示一个恢复点（由recovery_frame_cnt等语法元素指示），在该点之后，从随机存取点开始解码的图片将在内容上正确或大致正确。这与目标专利中恢复点SEI消息“指示在所述一个图片之前或所述一个图片处的随机存取点处开始解码之后作为恢复点的随后图片”的作用完全相同。结合特征A的论述，当恢复点SEI消息通过视图ID与特定图片关联时，其指示的恢复点功能同样适用于该视图/层。因此，该技术特征被对比文件直接公开。 |
| **技术特征F**：在所述恢复点处所述多层位流的解码将在内容上正确或大致正确。<br>**判断结果**：直接公开(<<<F>>>) | [0256] “A recovery point is defined as a point in the bitstream at which the recovery of an exact or an approximate representation of the decoded pictures represented by the bitstream may be achieved after a random access or broken link at the preceding random access point.”<br>[0260] “All decoded pictures at or subsequent to a recovery point in output order are correct or approximately correct in content.” | 对比文件明确定义了恢复点的作用，即在该点及之后（按输出顺序），解码出的图片在内容上是正确或大致正确的。这完全对应于目标专利特征F的描述。因此，该技术特征被对比文件直接公开。 |
| **技术特征G**：所述随后图片与所述一个图片包含在所述多层位流的同一层中。<br>**判断结果**：隐含公开(<<<g>>>) | [0320] “When a recovery point SEI message is contained in an MVC scalable nesting SEI message, the semantics of the contained recovery point SEI message may be interpreted in such a manner that decoding of only the views indicated in the MVC scalable nesting SEI message are required to recovery the indicated views...”<br>[0273] “VRA pictures enable starting the decoding of a subset of the views present in the bitstream.” | 对比文件描述了视图随机访问（VRA）图片允许从位流中存在的视图子集开始解码。当恢复点SEI消息通过MVC可缩放嵌套SEI消息与特定视图关联时，其指示的恢复过程是针对这些“指示的视图”的。本领域技术人员可以理解，恢复过程（即从随机存取点到恢复点）所涉及的解码图片序列，自然发生在所关联的同一视图（层）之内。因为恢复点SEI消息的目的是为该特定视图提供恢复信息，其指示的“随后图片”必然属于该同一视图，否则恢复信息将无意义。因此，该技术特征被对比文件隐含公开。 |
| **技术特征H**：以及至少部分地基于所述恢复点SEI消息与所述一个图片的所述关联而译码所述视频数据。<br>**判断结果**：隐含公开(<<<h>>>) | [0275] “An encoder 300 according to some example embodiments may operate as follows and as depicted in the flow diagram of FIG. 8. ... The encoder 300 determines that a certain access unit is a random access point for the bitstream e.g. based on a pre-determined random access point frequency and that this certain access unit is to be coded as a VRA access unit (block 802 in FIG. 8).”<br>[0290] “A decoder 400 according to some example embodiments may operate as follows and as depicted in the flow diagram of FIG. 9. ... The decoder 400 detects 902 a view random access (VRA) access unit using one of the indications in the bitstream, packet encapsulation format, or container file format. The decoder then concludes 904, which views can be decoded starting from the detected VRA access unit. ... The decoder starts the decoding of the bitstream from the detected VRA access unit (block 906 in FIG. 9).” | 对比文件详细描述了编码器和解码器如何利用视图随机访问（VRA）访问单元（其指示包含恢复点SEI消息等信息）来进行视频数据的编码和解码（译码）。编码器确定并编码VRA访问单元，解码器检测VRA访问单元并基于此开始解码特定的视图。这个过程的本质就是“至少部分地基于”恢复点SEI消息（或同等功能的指示）与特定视图（进而与特定图片）的关联来指导视频数据的译码操作。虽然对比文件主要使用“VRA访问单元”这一术语，但其实现机制依赖于如恢复点SEI消息等语法元素与视图的关联。因此，本领域技术人员可以合理推断出目标专利特征H所描述的技术方案。该技术特征被对比文件隐含公开。 |
<<<B>>><<<D>>><<<E>>><<<F>>><<<a>>><<<c>>><<<g>>><<<h>>>