公开(公告)号：US20220020167A1

公开(公告)日：2022-01-20

申请号：US17132054

申请日：2020-12-23

申请人： HYUNDAI MOTOR COMPANY ,  KIA MOTORS CORPORATION ,  KOREA ADVANCED INSTITUTE OF SCIENCE AND TECHNOLOGY

发明人： Ha Yong WOO ,  Jun Mo KIM ,  Beom Young KIM ,  Si Haeng LEE ,  Jang Hyeon LEE ,  Do Yeon KIM

IPC分类号： G06T7/73 ,  G06T11/20 ,  G06K9/32

摘要： A method for detecting an anchor-free object based on deep learning may include receiving a target image for object detection, inputting the image to a backbond network and performing learning to generate a heatmap, post-processing the heatmap, and outputting the post-processed result on a screen.

公开(公告)号：US20190294907A1

公开(公告)日：2019-09-26

申请号：US16437495

申请日：2019-06-11

申请人： Samsung Electronics Co., Ltd. ,  Korea Advanced Institute of Science and Technology

发明人： Byungin YOO ,  Jun Ho YIM ,  Jun Mo KIM ,  Hee Chul JUNG ,  Chang Kyu CHOI ,  Jaejoon HAN

IPC分类号： G06K9/20 ,  G06K9/62 ,  G06N3/08 ,  G06N3/04 ,  G06T5/00 ,  G06T15/20

摘要： At least some example embodiments disclose a device and a method for generating a synthetic image and a different-angled image and eliminating noise. The method may include receiving input images, extracting feature values corresponding to the input images using an image learning model, the image learning model permitting an input and an output to be identical and generating a synthetic image based on the feature values corresponding to the input images using the image learning model.

公开(公告)号：US20160277732A1

公开(公告)日：2016-09-22

申请号：US14768836

申请日：2015-04-30

申请人： KOREA ADVANCED INSTITUTE OF SCIENCE AND TECHNOLOGY

发明人： Jun Mo KIM ,  Hee Chul JUNG

IPC分类号： H04N19/105 ,  H04N19/139 ,  H04N19/154 ,  H04N19/895 ,  H04N19/37 ,  H04N19/56 ,  H04N19/137 ,  H04N19/119 ,  H04N19/167

CPC分类号： G06T7/11 ,  G06K9/00744 ,  G06K9/4642 ,  G06K9/6215 ,  G06K9/6231 ,  G06K2009/6295 ,  G06T7/143 ,  G06T7/162 ,  G06T7/215 ,  G06T7/246 ,  G06T7/33 ,  G06T7/73 ,  G06T2207/10016 ,  H04N19/105 ,  H04N19/137 ,  H04N19/139 ,  H04N19/154 ,  H04N19/521 ,  H04N19/54 ,  H04N19/543 ,  H04N19/895

摘要： The video-motion segmentation method using a randomized voting is provided which includes receiving the video, extracting a plurality of feature points from the video, and grouping the plurality of feature points by applying a randomized voting method using a score histogram on each of the at least some feature points of the plurality of feature points.

摘要翻译： 提供了一种使用随机投票的视频运动分割方法，其包括接收视频，从视频中提取多个特征点，并且通过应用随机投票方法对所述多个特征点进行分组，该方法使用每个特征点 至少一些特征点的多个特征点。