-
公开(公告)号：US20170213385A1
公开(公告)日：2017-07-27
申请号：US15405728
申请日：2017-01-13
Inventor： Jung Jae YU , Hyun KANG , Jae Hean KIM , Hye Sun KIM , Chang Joon PARK , Yun Ji BAN , Dong Wan RYOO , Jin Sung CHOI
CPC classification number: G06T17/20 , G06K9/00221 , G06K9/00268 , G06K9/4604 , G06K2209/40 , G06T19/20 , H04N7/185
Abstract: An apparatus for generating a 3-dimensional face model includes a multi-view image capturer configured to sense a motion of the mobile device and automatically capture still images from two or more directions; and a 3D model generator configured to generate a 3D face mode using the two or more still images obtained by the multi-view image capturer.
-
公开(公告)号：US20190164351A1
公开(公告)日：2019-05-30
申请号：US16057664
申请日：2018-08-07
Inventor： Hye Sun KIM , Kyung Kyu KANG , Chang Joon PARK , Yun Ji BAN , Dong Wan RYOO , Jung Jae YU , Man Hee LEE , Chang Woo CHU
CPC classification number: G06T19/20 , G06T15/04 , G06T17/205 , G06T2200/08 , G06T2219/2012
Abstract: Disclosed is a method of reconstructing a three-dimensional color mesh and an apparatus for the same. According to an embodiment of the present disclosure, the method includes: receiving mesh information of an object, multiple multi-view images obtained by photographing the object at different positions, and camera parameter information corresponding to the multiple multi-view images; constructing a texture map with respect to the object on the basis of the received information and setting a texture patch referring to a color value of the same multi-view image; correcting a color value of a vertex included for each texture patch; and performing rendering with respect to the object by applying the corrected color value of the vertex to the texture map.
-
3.发明申请公开(公告)号：US20130266212A1
公开(公告)日：2013-10-10
申请号：US13859634
申请日：2013-04-09
Inventor： Hye-Sun KIM , Yun Ji BAN , Kyung Ho JANG , Jung Jae YU , Myung Ha KIM , Ho Wook JANG , Seung Woo NAM , Hae Dong KIM
IPC: H04N13/00
CPC classification number: H04N13/128 , H04N13/261 , H04N13/271
Abstract: A method of converting a two-dimensional video to a three-dimensional video, the method comprising: comparing an image of an nth frame with an accumulated image until an n−1th frame in the two-dimensional video to calculate a difference in a color value for each pixel; generating a difference image including information on a change in a color value for each pixel of the nth frame; storing an accumulated image until the nth frame by accumulating the information on the change in the color value for each pixel until the nth frame; performing an operation for a pixel in which a change in a color value is equal to or larger than a predetermined level by using the difference image to generate a division image and a depth map image; and converting the image of the nth frame to a three-dimensional image by using the depth map image.
Abstract translation: 一种将二维视频转换为三维视频的方法，所述方法包括：将第n帧的图像与累积图像进行比较，直到二维视频中的第n-1帧，以计算颜色差异 每个像素的值; 生成包括关于第n帧的每个像素的颜色值的变化的信息的差分图像; 通过累积关于每个像素的颜色值的变化的信息直到第n帧来存储累积图像直到第n帧; 通过使用差分图像对颜色值的变化等于或大于预定水平的像素进行操作以产生分割图像和深度图图像; 以及通过使用深度图图像将第n帧的图像转换为三维图像。
-
公开(公告)号：US20240177473A1
公开(公告)日：2024-05-30
申请号：US18496063
申请日：2023-10-27
Inventor： Jaehwan KIM , Jung Jae YU , Wonyoung YOO , Juwon LEE
CPC classification number: G06V10/82 , G06T7/75 , G06V10/7715 , G06T2207/30196
Abstract: A neural network device for learning dependency of feature data includes: a memory in which at least one program is stored; and a processor that performs a calculation by executing the at least one program, in which the processor is configured to acquire graph information including a data node for a human body; extract feature data corresponding to a plurality of joints constituting the human body from the graph information; acquire a self-attention output corresponding to the feature data based on a self-attention mechanism; and generate result data for a motion of the human body based on the self-attention output, and the self-attention output includes position information acquired based on positional encoding of the feature data and structural information acquired based on geodesic encoding of the feature data.
-
公开(公告)号：US20230237629A1
公开(公告)日：2023-07-27
申请号：US18077434
申请日：2022-12-08
Inventor： Juwon LEE , Jung Jae YU , Wonyoung YOO
CPC classification number: G06T5/50 , G06V40/171 , G06T7/11 , G06T3/0068 , G06T3/0093 , G06T2207/30201 , G06T2207/20221
Abstract: A method and apparatus for generating a face-harmonized image are disclosed. The method of generating a face-harmonized image includes receiving an input image, extracting facial landmarks from a target image and the input image, generating a face-removed image of the target image based on a facial mask region, extracting a user face image from the input image, transforming the user face image to correspond to the facial mask region, generating a face-blended image by blending the transformed user face image with the target image, extracting a feature map of the face-blended image, generating a combined feature map based on the feature map of the face-blended image and a feature map of the target image, generating a face harmonization result image based on the combined feature map, and providing the generated face harmonization result image.
-
Patent Agency Ranking
公开(公告)号：US20230162409A1
公开(公告)日：2023-05-25
申请号：US17739375
申请日：2022-05-09
Inventor： Jung Jae YU , Ju Won LEE , Won Young YOO
CPC classification number: G06T11/001 , G06T7/11 , G06T3/40 , G06T2207/20084 , G06T2207/10024 , G06T2207/20081
Abstract: Provided are a system and method for generating an image implemented in the same style as a style of a single image sample based on a result of learning the single image sample, the system including: an image learning unit configured to train a deep learning network model based on an image sample of a specific style to generate a same style generative model network; and an image generating unit configured to input a layout image to the same style generative model network to generate an image having the same style as the specific style image sample while conforming to a layout structure of the layout image.
-
公开(公告)号：US20220222525A1
公开(公告)日：2022-07-14
申请号：US17554870
申请日：2021-12-17
Inventor： Su Woong LEE , Seungjae LEE , Jong-Gook KO , Wonyoung YOO , Jung Jae YU , Keun Dong LEE , Yongsik LEE , Da-Un JUNG
Abstract: Provided are a method and system for training a dynamic deep neural network. The method for training a dynamic deep neural network includes receiving an output of a last layer of the deep neural network and outputting a first loss, receiving an output of a routing module according to an input class of the deep neural network and outputting a second loss, calculating a third loss based on the first loss and the second loss, and updating a weight of the deep neural network by using the third loss.
-
8.发明申请APPARATUS AND METHOD FOR PRODUCING STEREOSCOPIC SUBTITLES BY ANALYZING THREE-DIMENSIONAL (3D) SPACE 审中-公开通过分析三维（3D）空间产生立体子囊的装置和方法公开(公告)号：US20140293019A1
公开(公告)日：2014-10-02
申请号：US14161932
申请日：2014-01-23
Inventor： Kyung Ho JANG , Hae Dong KIM , Ho Wook JANG , Seung Woo NAM , Jung Jae YU , Myung Ha KIM , Su Ran PARK
IPC: H04N13/00
CPC classification number: H04N13/183 , G06T5/005 , G06T2207/10021 , G06T2207/30244 , H04N13/111 , H04N13/156 , H04N13/239
Abstract: An apparatus and method of producing stereoscopic subtitles by analyzing a three-dimensional (3D) space is disclosed, the apparatus including a camera position calculator to calculate a position of a first camera and a position of a second camera from a first image and a second image, respectively, a subtitle flat arranger to arrange a subtitle flat using a viewing direction of the first camera and a viewing direction of the second camera at the calculated positions, and a subtitle producer to produce subtitles using the subtitle flat.
Abstract translation: 公开了一种通过分析三维（3D）空间来产生立体字幕的装置和方法，所述装置包括：相机位置计算器，用于计算第一相机的位置和第一相机的位置;以及第二 图像分别是使用第一相机的观看方向来排列小标题平面的小标题平面编排器和在所计算的位置处的第二相机的观看方向，以及字幕制作器，以使用字幕平面来产生字幕。
-
公开(公告)号：US20130236099A1
公开(公告)日：2013-09-12
申请号：US13791791
申请日：2013-03-08
Inventor： Jung Jae YU , Ho Wook Jang , Seung Woo Nam , Kyung Ho Jang , Hae Dong Kim , Hye-Sun Kim , Yun Ji Ban , Myung Ha Kim
IPC: G06T7/00
CPC classification number: G06T7/0097 , G06T7/248 , G06T2207/10016 , G06T2207/10024 , G06T2207/20036 , G06T2207/30196
Abstract: Disclosed are an apparatus and a method for extracting a foreground layer from an image sequence that extract a foreground object layer area in which a depth value is discontinuous with that of a background from an input image sequence. By using the present disclosure, the layer area is automatically tracked in the subsequent frames through user's setting in the start frame in the image sequence in which the depth values of the foreground and the background are discontinuous, thereby extracting the foreground layer area in which the drift phenomenon and the flickering phenomenon are reduced.
Abstract translation: 公开了一种从从输入图像序列中提取深度值与背景的深度值不连续的前景对象层区域的图像序列中提取前景层的装置和方法。 通过使用本公开，通过在前景和背景的深度值不连续的图像序列中的开始帧中的用户设置，在随后的帧中自动跟踪层区域，从而提取其中 漂移现象和闪烁现象减少。
-
公开(公告)号：US20240096020A1
公开(公告)日：2024-03-21
申请号：US18468162
申请日：2023-09-15
Inventor： Jung Jae YU , Jae Hwan KIM , Ju Won LEE , Won Young YOO
Abstract: A method of generating a moving viewpoint motion picture, which is performed by a processor that executes at least one instruction stored in a memory, may comprise: obtaining an input image; generating a trimap from the input image; generating a depth map using the input image; generating a foreground mesh/texture map model based on a foreground alpha map obtained based on the trimap and foreground depth information obtained based on the trimap and the depth map; and generating a moving viewpoint motion picture based on the foreground mesh/texture map model.
-
-
-
-
-
-
-
-
-
Search Results
14
records
(took 0.028 seconds)
