-
公开(公告)号：US20230153066A1
公开(公告)日：2023-05-18
申请号：US17702242
申请日：2022-03-23
Inventor： Jong-Ryul LEE , Yong-Hyuk MOON , Yong-Ju LEE
CPC classification number: G06F7/49947 , G06F7/50 , G06F7/523 , G06N3/04
Abstract: Disclosed herein is a method for measuring the weight of a discrete entity, performed in a neural network model configured with multiple layers, the method including receiving data configured with the indices of discrete entities, converting the data into embedding vectors corresponding to respective indices through an embedding layer, generating a masked vector through element-wise multiplication between a mask vector and the embedding vector, calculating a loss using output based on the masked vector, and training the model based on the loss.
-
2.发明申请APPARATUS AND METHOD FOR CODING AND DECODING MULTI-OBJECT AUDIO SIGNAL WITH VARIOUS CHANNEL INCLUDING INFORMATION BITSTREAM CONVERSION 审中-公开用于编码和解码多个通道的多目标音频信号的装置和方法，包括信息比特转换公开(公告)号：US20130132098A1
公开(公告)日：2013-05-23
申请号：US13732682
申请日：2013-01-02
Inventor： Seung-Kwon BEACK , Jeong-Il SEO , Tae-Jin LEE , Yong-Ju LEE , Dae-Young JANG , Jin-Woo HONG , Jin-Woong KIM , Kyeong-Ok KANG
IPC: G10L19/008
CPC classification number: G10L19/008 , G10L19/0017 , G10L19/173 , H04S3/002 , H04S7/30 , H04S2400/11 , H04S2420/03
Abstract: Provided is an apparatus and method for coding and decoding multi-object audio signals with various channels and providing backward compatibility with a conventional spatial audio coding (SAC) bitstream. The apparatus includes: an audio object coding unit for coding audio-object signals inputted to the coding apparatus based on a spatial cue and creating rendering information for the coded audio-object signals, where the rendering information provides a coding apparatus including spatial cue information for audio-object signals; channel information of the audio-object signals; and identification information of the audio-object signals, and used in coding and decoding of the audio signals.
Abstract translation: 提供了一种用于利用各种信道对多对象音频信号进行编码和解码并提供与常规空间音频编码（SAC）比特流的向后兼容性的装置和方法。 该装置包括：音频对象编码单元，用于基于空间提示对输入到编码装置的音频对象信号进行编码，并创建用于编码的音频对象信号的再现信息，其中，所述再现信息提供包括用于 音频对象信号; 音频对象信号的信道信息; 以及音频对象信号的识别信息，并用于音频信号的编码和解码。
-
3.发明公开公开(公告)号：US20240202999A1
公开(公告)日：2024-06-20
申请号：US18480177
申请日：2023-10-03
Inventor： Yong-Ju LEE
CPC classification number: G06T11/206 , G06V20/70 , G06V2201/07
Abstract: Disclosed herein is a method for organizing and merging scene understanding information of an AI agent. The method includes acquiring an image of a first space, recognizing objects in the image, structuring information about the relationship between the object and information about the states of the objects in the form of a graph, and merging the structured information with information received from a nearby AI agent.
-
4.发明申请公开(公告)号：US20140229506A1
公开(公告)日：2014-08-14
申请号：US14017476
申请日：2013-09-04
Inventor： Yong-Ju LEE
IPC: G06F17/30
CPC classification number: G06F16/24535 , G06F16/245 , G06F16/24554 , G06F16/24568
Abstract: Disclosed herein is a data stream processing apparatus and method using query partitioning, which allow data stream processing apparatuses to perform partitioned processing/parallel processing on partitioned sub-queries. The proposed data stream processing apparatus using query partitioning receives a query from a user, partitions the query into a plurality of sub-queries, transmits the partitioned sub-queries to another data stream processing apparatus or a sub-query processing unit, integrates the results of the processing of sub-queries processed by the other data stream processing apparatus and the sub-query processing unit with each other, generates a response to the query, and transmits the generated response to the user.
Abstract translation: 这里公开了一种使用查询分区的数据流处理装置和方法，其允许数据流处理装置对分区子查询执行分区处理/并行处理。 所提出的使用查询分区的数据流处理装置从用户接收查询，将查询分割成多个子查询，将分割后的子查询发送到另一数据流处理装置或子查询处理单元， 由其他数据流处理装置和子查询处理单元处理的子查询的处理彼此产生对查询的响应，并将生成的响应发送给用户。
-
5.发明公开公开(公告)号：US20240176653A1
公开(公告)日：2024-05-30
申请号：US18311792
申请日：2023-05-03
Inventor： Yong-Ju LEE
CPC classification number: G06F9/4881 , G06F8/30
Abstract: Disclosed herein is a method for task planning for collaboration of artificial intelligence (AI) agents. The method includes generating a scene graph using an image acquired by an AI agent and a human instruction and generating a machine instruction set for objects in the scene graph, and the scene graph includes relevance information between each of the objects in the scene graph and the human instruction.
-
Patent Agency Ranking
6.发明申请公开(公告)号：US20190215633A1
公开(公告)日：2019-07-11
申请号：US16356410
申请日：2019-03-18
Inventor： Seung-Kwon BEACK , Jeong-Il SEO , Dae-Young JANG , Tae-Jin LEE , Yong-Ju LEE , Jin-Woo HONG , Jin-Woong KIM
IPC: H04S7/00 , G10L19/008
CPC classification number: H04S7/30 , G10L19/008 , H04S2400/11
Abstract: The present research relates to controlling rendering of multi-object or multi-channel audio signals. The present research provides a method and apparatus for controlling rendering of multi-object or multi-channel audio signals based on spatial cues in a process of decoding the multi-object or multi-channel audio signals. To achieve the purpose, the method suggested in the research controls rendering in a spatial cue domain in the process of decoding the multi-object or multi-channel audio signals.
-
公开(公告)号：US20240420392A1
公开(公告)日：2024-12-19
申请号：US18612220
申请日：2024-03-21
Inventor： Sang-Hun JEON , Yong-Ju LEE
IPC: G06T11/20 , G06T3/18 , G06T7/12 , G06V10/25 , G06V10/764 , G06V10/771 , G06V10/80
Abstract: Disclosed herein is an apparatus and method for scene graph generation. The apparatus may include a backbone network for extracting a first feature map from an input image, an encoder for extracting a second feature map that is based on a mask for the shape of an object within a bounding box using the first feature map and generating a third feature map by combining the first feature map and the second feature map, and a decoder for generating a scene graph by predicting the relationship between objects from the third feature map.
-
8.发明申请公开(公告)号：US20200267488A1
公开(公告)日：2020-08-20
申请号：US16869902
申请日：2020-05-08
Inventor： Seung-Kwon BEACK , Jeong-Il SEO , Dae-Young JANG , Tae-Jin LEE , Yong-Ju LEE , Jin-Woo HONG , Jin-Woong KIM
IPC: H04S7/00 , G10L19/008
Abstract: The present research relates to controlling rendering of multi-object or multi-channel audio signals. The present research provides a method and apparatus for controlling rendering of multi-object or multi-channel audio signals based on spatial cues in a process of decoding the multi-object or multi-channel audio signals. To achieve the purpose, the method suggested in the research controls rendering in a spatial cue domain in the process of decoding the multi-object or multi-channel audio signals.
-
9.发明申请METHOD AND APPARATUS FOR GENERATING SIDE INFORMATION BITSTREAM OF MULTI-OBJECT AUDIO SIGNAL 审中-公开用于生成多个对象音频信号的侧面信息比特的方法和装置公开(公告)号：US20160165375A1
公开(公告)日：2016-06-09
申请号：US15041209
申请日：2016-02-11
Inventor： Jeong-Il SEO , Seung-Kwon BEACK , Tae-Jin LEE , Yong-Ju LEE , Dae-Young JANG , Kyeongok KANG , Jin-Woo HONG , Jin-Woong KIM , Chieteuk AHN
IPC: H04S7/00 , G10L19/008 , H04S5/00
CPC classification number: H04S7/308 , G10L19/008 , H04S5/00 , H04S2400/03 , H04S2400/11
Abstract: Provided is a method and apparatus for generating a side information bitstream of a multi-object audio signal. The apparatus for generating a side information bitstream of a multi-object audio signal includes a spatial cue information input unit configured to receive spatial cue information generated in an encoder of the multi-object audio signal, a preset information input unit configured to receive preset information for the multi-object audio signal, and a side information bitstream generator configured to generate the side information bitstream based on the spatial cue information and the preset information. The side information bitstream includes a header region and a frame region, and the preset information is included in the frame region.
Abstract translation: 提供了一种用于生成多对象音频信号的侧信息比特流的方法和装置。 用于生成多对象音频信号的侧信息比特流的装置包括：空间提示信息输入单元，被配置为接收在多对象音频信号的编码器中生成的空间提示信息;预设信息输入单元，被配置为接收预设信息 以及侧信息比特流发生器，被配置为基于空间提示信息和预设信息生成侧信息比特流。 侧信息比特流包括报头区域和帧区域，并且预设信息被包括在帧区域中。
-
10.发明申请OBJECT-BASED 3-DIMENSIONAL AUDIO SERVICE SYSTEM USING PRESET AUDIO SCENES 审中-公开使用预置音频场景的基于对象的三维音频服务系统公开(公告)号：US20140185810A1
公开(公告)日：2014-07-03
申请号：US14199188
申请日：2014-03-06
Inventor： Yong-Ju LEE , Tae-Jin LEE , Jae-Hyoun YOO , Kyeong-Ok KANG , Jin-Woo HONG , In-Seon JANG , Jeong-Il SEO , Dae-Young JANG
IPC: H04S3/00
CPC classification number: H04S3/008 , H04H20/47 , H04H20/88 , H04H20/89 , H04H40/36 , H04H60/04 , H04N21/44012 , H04N21/8106
Abstract: Provided are an object-based three dimensional (3-D) audio service system using preset audio scenes and a method thereof. The system and the method are suggested for enabling a user to easily and conveniently watch and listen an object based 3-D audio service by eliminating inconvenience that requires a user to control each of object audio signals of sound sources. The system includes: audio input means for inputting an audio signal; preset audio scene generating means for extracting object audio signals from the audio signal inputted through the audio input means and generating more than one of 3-D audio scene information by arranging the extracted object audio signals in a 3-D space and editing features of each object; and encoding means for encoding and multiplexing the audio signal and the 3-D audio scene information for each object audio signal.
Abstract translation: 提供了使用预设音频场景的基于对象的三维（3-D）音频服务系统及其方法。 建议使用该系统和方法，以便用户能够通过消除需要用户控制声源的每个对象音频信号的不便，来容易且方便地观看和收听基于对象的3-D音频服务。 该系统包括：用于输入音频信号的音频输入装置; 预置音频场景产生装置，用于从通过音频输入装置输入的音频信号中提取对象音频信号，并通过将提取的对象音频信号排列在3-D空间中并产生多个3-D音频场景信息，并且编辑每个 目的; 以及用于对每个对象音频信号对音频信号和3-D音频场景信息进行编码和多路复用的编码装置。
-
-
-
-
-
-
-
-
-
Search Results
12
records
(took 0.02 seconds)
