公开(公告)号：EP3467711A1

公开(公告)日：2019-04-10

申请号：EP18192815.1

申请日：2018-09-05

申请人： StradVision, Inc.

发明人： The designation of the inventor has not yet been filed

IPC分类号： G06K9/46 ,  G06K9/62

摘要： A learning method for improving image segmentation including steps of: (a) acquiring a (1-1)-th to a (1-K)-th feature maps through an encoding layer if a training image is obtained; (b) acquiring a (3-1)-th to a (3-H)-th feature maps by respectively inputting each output of the H encoding filters to a (3-1)-th to a (3-H)-th filters; (c) performing a process of sequentially acquiring a (2-K)-th to a (2-1)-th feature maps either by (i) allowing the respective H decoding filters to respectively use both the (3-1)-th to the (3-H)-th feature maps and feature maps obtained from respective previous decoding filters of the respective H decoding filters or by (ii) allowing respective K-H decoding filters that are not associated with the (3-1)-th to the (3-H)-th filters to use feature maps gained from respective previous decoding filters of the respective K-H decoding filters; and (d) adjusting parameters of CNN.

公开(公告)号：EP3913530A1

公开(公告)日：2021-11-24

申请号：EP21172016.4

申请日：2021-05-04

申请人： Stradvision, Inc.

发明人： RYU, Wooju ,  JE, Hongmo ,  KANG, Bongnam ,  KIM, Yongjoong

IPC分类号： G06K9/00 ,  G06K9/62

摘要： A method for updating an object detector of an autonomous vehicle to adapt the object detector to a driving circumstance is provided. The method includes steps of: a learning device (a) (i) inputting a training image, corresponding to a driving circumstance, into a circumstance-specific object detector to apply (i-1) convolution to the training image to generate a circumstance-specific feature map, (i-2) ROI pooling to the circumstance-specific feature map to generate a circumstance-specific pooled feature map, and (i-3) fully-connected operation to the circumstance-specific pooled feature map to generate circumstance-specific object detection information and (ii) inputting the circumstance-specific feature map into a circumstance-specific ranking network to (ii-1) apply deconvolution to the circumstance-specific feature map and generate a circumstance-specific segmentation map and (ii-2) generate a circumstance-specific rank score via a circumstance-specific discriminator; and (b) training the circumstance-specific object detector, the circumstance-specific deconvolutional layer, the circumstance-specific convolutional layer, and the circumstance-specific discriminator.

公开(公告)号：EP3907654A1

公开(公告)日：2021-11-10

申请号：EP21153164.5

申请日：2021-01-25

申请人： Stradvision, Inc.

发明人： KIM, Kye-Hyeon ,  GWEON, Sung An ,  KIM, Yongjoong ,  KANG, Bongnam

IPC分类号： G06K9/00 ,  G06K9/32 ,  G06K9/34 ,  G06K9/46 ,  G06K9/62 ,  G06N3/04 ,  G06N3/08

摘要： Processes of explainable active learning, for an object detector, by using a Bayesian dual encoder is provided. The processes include: (a) inputting test images into the object detector to generate cropped images, resizing the test images and the cropped images, and inputting the resized images into a data encoder to output data codes; (b) (b1) one of (i) inputting the test images into the object detector, applying Bayesian output embedding and resizing the activation entropy maps and the cropped activation entropy maps, and (ii) inputting resized object images and applying the Bayesian output embedding and (b2) inputting the resized activation entropy maps into a model encoder to output model codes; and (c) (i) confirming reference data codes, selecting specific test images as rare samples, and updating the data codebook, and (ii) confirming reference model codes and selecting specific test images as hard samples.

公开(公告)号：EP3885998A1

公开(公告)日：2021-09-29

申请号：EP20212334.5

申请日：2020-12-08

申请人： Stradvision, Inc.

发明人： KIM, Kye-Hyeon ,  JE, Hongmo ,  KANG, Bongnam ,  RYU, Wooju

IPC分类号： G06N3/04 ,  G06N3/08

摘要： A method for training a deep learning network based on artificial intelligence is provided. The method includes steps of: a learning device (a) inputting unlabeled data into an active learning network to acquire sub unlabeled data and inputting the sub unlabeled data into an auto labeling network to generate new labeled data; (b) allowing a continual learning network to sample the new labeled data and existing labeled data to generate a mini-batch, and train the existing learning network using the mini-batch to acquire a trained learning network, wherein part of the mini-batch are selected by referring to specific existing losses; and (c) (i) allowing an explainable analysis network to generate insightful results on validation data and transmit the insightful results to a human engineer to transmit an analysis of the trained learning network and (ii) modifying at least one of the active learning network and the continual learning network.

公开(公告)号：EP3731143A1

公开(公告)日：2020-10-28

申请号：EP20151765.3

申请日：2020-01-14

申请人： StradVision, Inc.

发明人： KIM, Kye-Hyeon ,  KIM, Yongjoong ,  KIM, Insu ,  KIM, Hak-Kyoung ,  NAM, Woonhyun ,  BOO, SukHoon ,  SUNG, Myungchul ,  YEO, Donghun ,  RYU, Wooju ,  JANG, Taewoong ,  JEONG, Kyungjoong ,  JE, Hongmo ,  CHO, Hojin

IPC分类号： G06K9/32 ,  G06K9/00 ,  G06K9/62 ,  G06K9/46

摘要： A method for economizing computing resources and verifying an integrity of parameters of a neural network by inserting test pattern into a background area of an input image is provided for fault tolerance, fluctuation robustness in extreme situations, functional safety on the neural network, and an annotation cost reduction. The method includes steps of: a computing device (a) generating t-th background prediction information of a t-th image by referring to information on each of a (t-2)-th image and a (t-1)-th image; (b) inserting the test pattern into the t-th image by referring to the t-th background prediction information, to thereby generate an input for verification; (c) generating an output for verification from the input for verification; and (d) determining the integrity of the neural network by referring to the output for verification and an output for reference. According to the method, a data compression and a computation reduction are achieved.

公开(公告)号：EP3702960A1

公开(公告)日：2020-09-02

申请号：EP20152203.4

申请日：2020-01-16

申请人： StradVision, Inc.

发明人： KIM, Kye-Hyeon ,  KIM, Yongjoong ,  KIM, Hak-Kyoung ,  NAM, Woonhyun ,  BOO, SukHoon ,  SUNG, Myungchul ,  SHIN, Dongsoo ,  YEO, Donghun ,  RYU, Wooju ,  LEE, Myeong-Chun ,  LEE, Hyungsoo ,  JANG, Taewoong ,  JEONG, Kyungjoong ,  JE, Hongmo ,  CHO, Hojin

IPC分类号： G06K9/00

摘要： A method for managing a smart database which stores facial images for face recognition is provided. The method includes steps of: a managing device (a) counting specific facial images corresponding to a specific person in the smart database where new facial images are continuously stored, and determining whether a first counted value, representing a count of the specific facial images, satisfies a first set value; and (b) if the first counted value satisfies the first set value, inputting the specific facial images into a neural aggregation network, to generate quality scores of the specific facial images by aggregation of the specific facial images, and, if a second counted value, representing a count of specific quality scores among the quality scores from a highest during counting thereof, satisfies a second set value, deleting part of the specific facial images, corresponding to the uncounted quality scores, from the smart database.

公开(公告)号：EP3690817A1

公开(公告)日：2020-08-05

申请号：EP20153532.5

申请日：2020-01-24

申请人： StradVision, Inc.

发明人： Kim, Kye-Hyeon ,  Kim, Yongjoong ,  Kim, Hak-Kyoung ,  Nam, Woonhyun ,  Boo, SukHoon ,  Sung, Myungchul ,  Shin, Dongsoo ,  Yeo, Donghun ,  Ryu, Wooju ,  Lee, Myeong-Chun ,  Lee, Hyungsoo ,  Jang, Taewoong ,  Jeong, Kyungjoong ,  Je, Hongmo ,  Cho, Hojin

IPC分类号： G06T7/80

摘要： A method for enhancing an accuracy of object distance estimation based on a subject camera by performing pitch calibration of the subject camera more precisely with additional information acquired through V2V communication is provided. And the method includes steps of: (a) a computing device, performing (i) a process of instructing an initial pitch calibration module to apply a pitch calculation operation to the reference image, to thereby generate an initial estimated pitch, and (ii) a process of instructing an object detection network to apply a neural network operation to the reference image, to thereby generate reference object detection information; (b) the computing device instructing an adjusting pitch calibration module to (i) select a target object, (ii) calculate an estimated target height of the target object, (iii) calculate an error corresponding to the initial estimated pitch, and (iv) determine an adjusted estimated pitch on the subject camera by using the error.

公开(公告)号：EP3690809A1

公开(公告)日：2020-08-05

申请号：EP20152786.8

申请日：2020-01-21

申请人： StradVision, Inc.

发明人： KIM, Kye-Hyeon ,  KIM, Yongjoong ,  KIM, Hak-Kyoung ,  NAM, Woonhyun ,  BOO, SukHoon ,  SUNG, Myungchul ,  SHIN, Dongsoo ,  YEO, Donghun ,  RYU, Wooju ,  LEE, Myeong-Chun ,  LEE, Hyungsoo ,  JANG, Taewoong ,  JEONG, Kyungjoong ,  JE, Hongmo ,  CHO, Hojin

IPC分类号： G06T7/00 ,  G06N3/02

摘要： A method for generating safe clothing patterns for a human-like figure is provided. The method includes steps of: a safe clothing-pattern generating device, (a) after acquiring an image of the human-like figure, generating a specific clothing pattern having an initial value, inputting the specific clothing pattern and the image of the human-like figure into a clothing composition network, combining the specific clothing pattern with a clothing of the human-like figure to generate a composite image; (b) inputting the composite image into an image translation network, translating surrounding environment on the composite image to generate a translated image, and inputting the translated image into an object detector to output detection information on the human-like figure; and (c) instructing a 1-st loss layer to calculate losses by referring to the detection information and a GT corresponding to the image of the human-like figure, and updating the initial value by using the losses.

公开(公告)号：EP3690756A1

公开(公告)日：2020-08-05

申请号：EP20153262.9

申请日：2020-01-23

申请人： Stradvision, Inc.

发明人： KIM, Kye-Hyeon ,  KIM, Yongjoong ,  KIM, Hak-Kyoung ,  NAM, Woonhyun ,  BOO, SukHoon ,  SUNG, Myungchul ,  SHIN, Dongsoo ,  YEO, Donghun ,  RYU, Wooju ,  LEE, Myeong-Chun ,  LEE, Hyungsoo ,  JANG, Taewoong ,  JEONG, Kyungjoong ,  JE, Hongmo ,  CHO, Hojin

IPC分类号： G06N3/04 ,  G06N3/08 ,  G06T7/00

摘要： A learning method for selecting specific information, to be used for updating an HD Map is provided. And the method includes steps of: (a) a learning device instructing a coordinate neural network to generate a local feature map and a global feature vector by applying a coordinate neural network operation to a coordinate matrix; (b) the learning device instructing a determination neural network to generate a first estimated suitability score to an N-th estimated suitability score by applying a determination neural network operation to the integrated feature map; (c) the learning device instructing a loss layer to generate a loss by referring to (i) the first estimated suitability score to the N-th estimated suitability score and (ii) a first Ground Truth(GT) suitability score to an N-th GT suitability score, and perform backpropagation by using the loss, to thereby learn parameters of the determination neural network and the coordinate neural network.

公开(公告)号：EP3690755A1

公开(公告)日：2020-08-05

申请号：EP20152464.2

申请日：2020-01-17

申请人： StradVision, Inc.

发明人： Kim, Kye-Hyeon ,  Kim, Yongjoong ,  Kim, Insu ,  Kim, Hak-Kyoung ,  Nam, Woonhyun ,  Boo, SukHoon ,  Sung, Myungchul ,  Yeo, Donghun ,  Ryu, Wooju ,  Jang, Taewoong ,  Jeong, Kyungjoong ,  Je, Hongmo ,  Cho, Hojin

IPC分类号： G06N3/04 ,  G06N3/08 ,  G06N7/00

摘要： A method for on-device continual learning of a neural network which analyzes input data is provided for smartphones, drones, vessels, or a military purpose. The method includes steps of: a learning device, (a) uniform-sampling new data to have a first volume, instructing a boosting network to convert a k-dimension random vector into a k-dimension modified vector, instructing an original data generator network to repeat outputting synthetic previous data of a second volume corresponding to the k-dimension modified vector and previous data having been used for learning, and generating a batch for a current-learning; and (b) instructing the neural network to generate output information corresponding to the batch. The method can be used for preventing catastrophic forgetting and an invasion of privacy, and for optimizing resources such as storage and sampling processes for training images. Further the method can be performed through a learning for Generative adversarial networks (GANs).