-
公开(公告)号：US20240193727A1
公开(公告)日：2024-06-13
申请号：US18332939
申请日：2023-06-12
发明人： Hyeonseung YU , Nahyup KANG , Hanjun KIM , Jaeyoung MOON , Juyoung LEE , Hwiryong JUNG , Inwoo HA , Seokpyo HONG
IPC分类号： G06T3/40 , G06T3/00 , G06T7/50 , G06T7/90 , G06V10/764
CPC分类号： G06T3/4053 , G06T3/0093 , G06T7/50 , G06T7/90 , G06V10/764 , G06T2207/10024 , G06V2201/07
摘要： A processor implemented method includes generating a semantic map indicating a visualization property assigned to an object of an obtained frame image having a first resolution and generating a reconstruction image using an image reconstruction machine learning model provided input based on the obtained frame image and the semantic map having a second resolution and including a second object having a visualization property indicated by the semantic map.
-
公开(公告)号：US20230297487A1
公开(公告)日：2023-09-21
申请号：US17889135
申请日：2022-08-16
发明人： Myeong Woo KIM , Kunwoo KIM , Changsu KIM , Hanjun KIM
CPC分类号： G06F11/3423 , G06F11/3037
摘要： A method and apparatus for estimating execution time of a neural network are provided, the method of estimating execution time of a neural network in a multi-core accelerator, the method including generating trace information including operation timing information for each core of the multi-core accelerator, and calculating the execution time of the neural network reflecting communication overhead between cores of the multi-core accelerator and memory access time for each core of the cores, based on the trace information.
-
公开(公告)号：US20240249383A1
公开(公告)日：2024-07-25
申请号：US18351963
申请日：2023-07-13
发明人： Seokpyo HONG , Nahyup KANG , Hanjun KIM , Jaeyoung MOON , Hyeonseung YU , Juyoung LEE , Hwiryong JUNG , Inwoo HA
CPC分类号： G06T3/4053 , G06T3/0093 , G06T3/4046
摘要： A method includes: inserting new pixels between original pixels for each of maps included in a first geometry buffer (or G-buffer) generated from a frame, wherein the maps represent geometric information of a three-dimensional (3D) model of an object included in the frame; generating a second G-buffer by setting values of the new pixels using a motion vector map that may be one of the maps; generating a third G-buffer by combining, with the second G-buffer, a result of updating only values of pixels masked based on an output of a pixel masking neural network to which the second G-buffer may be input; generating a fourth G-buffer by updating values of pixels by inputting the third G-buffer to a G-buffer reconstruction neural network; and update, based on the fourth G-buffer, the resolution of a subsequent frame that follows the frame.
-
公开(公告)号：US20240169482A1
公开(公告)日：2024-05-23
申请号：US18323024
申请日：2023-05-24
发明人： Nahyup KANG , Hanjun KIM , Jaeyoung MOON , Hyeonseung YU , Juyoung LEE , Hwiryong JUNG , Inwoo HA , Seokpyo HONG
CPC分类号： G06T3/4046 , G06N3/045 , G06N3/08 , G06T3/0093 , G06T3/4007
摘要： An electronic device includes a memory for storing a first neural network and a second neural network including a plurality of residual blocks and an upscaling block, and a processor for selecting a residual block from among the plurality of residual blocks for an input patch image of a first frame based on the second neural network and generating an output patch image of the first frame by upscaling the input patch image of the first frame to an image having a target resolution based on the selected residual block and the upscaling block.
-
公开(公告)号：US20240161236A1
公开(公告)日：2024-05-16
申请号：US18316120
申请日：2023-05-11
发明人： Hwiryong JUNG , Nahyup KANG , Hanjun KIM , Jaeyoung MOON , Hyeonseung YU , Juyoung LEE , Inwoo HA , Seokpyo HONG
CPC分类号： G06T3/4053 , G06T7/248 , G06V10/7515
摘要： An adaptive super sampling method includes: rendering frames of a three-dimensional (3D) model, the frames including a current frame and a previous frame preceding the current frame; determining motion vectors indicating a correspondence relationship between pixels in the current frame and pixels in the previous frame; generating a geometric identifier maps (G-ID maps) respectively corresponding to the current frame and the previous frame based on 3D geometrical properties associated with the pixels in the current frame and the previous frame; based on the motion vectors and the G-ID maps, generating an artifact map predicting where artifacts will occur from inter-frame super sampling of the current frame; and performing adaptive super sampling on the current frame based on the artifact map.
-
公开(公告)号：US20240161233A1
公开(公告)日：2024-05-16
申请号：US18303210
申请日：2023-04-19
发明人： Jaeyoung MOON , Nahyup KANG , Hanjun KIM , Hyeonseung YU , Juyoung LEE , Hwiryong JUNG , Inwoo HA , Seokpyo HONG
CPC分类号： G06T3/4046 , G06T3/0093 , G06T5/002 , G06T5/005 , G06T5/50 , G06T7/20 , G06T2200/04 , G06T2207/20021 , G06T2207/20084
摘要： A supersampling method and apparatus are provided. The method includes: receiving a low-resolution three-dimensional (3D) image comprising a current frame and receiving a previous frame preceding the current frame; generating a low-resolution partial image by repeatedly sampling sub-pixel regions of the current frame; warping a high-resolution image, of the previous frame, which has been outputted from a neural network, to a current view corresponding to the current frame; replacing a partial region of the warped high-resolution image of the previous frame with image data from the low-resolution partial image; and generating a high-resolution image of the current frame by applying the high-resolution image of the previous frame, in which the partial region has been replaced, to the neural network.
-
公开(公告)号：US20190302445A1
公开(公告)日：2019-10-03
申请号：US16367010
申请日：2019-03-27
发明人： Seongjin CHO , Naekyung KIM , Yongjoon KIM , Hanjun KIM , Jihyuk LIM
摘要： A wearable device includes at least one lens, an output device, at least one sensor, and at least one processor. The at least one processor is configured to identify a level of light sensed through the at least one sensor. The at least one processor is also configured to provide a notification related to the level of the light by using the output device when the level of the light satisfies a preset condition.
-
8.发明申请公开(公告)号：US20220180150A1
公开(公告)日：2022-06-09
申请号：US17177909
申请日：2021-02-17
发明人： Yeongsik LEE , Youngsok KIM , Hanjun KIM , Sungjun CHO , Seonyeong HEO
摘要： A data processing method and apparatus using a neural network, and an electronic device including the data processing apparatus. The data processing method includes identifying an operator that selects one of a plurality of execution paths for a portion of the neural network while sequentially executing layers included in the neural network, selecting a specific execution path, from among the plurality of execution paths, based on a remaining time that is left for an inference of the neural network, and obtaining a result of the inference of the neural network through the specific execution path.
-
-
-
-
-
-
-
搜索结果
8 条记录 (耗时 0.016 秒)
