-
1.发明申请公开(公告)号：US20220292397A1
公开(公告)日：2022-09-15
申请号：US17633402
申请日：2019-08-21
Applicant: NEC Corporation
Inventor： Katsuhiko TAKAHASHI , Tetsuo INOSHITA , Asuka ISHII , Gaku NAKANO
IPC: G06N20/00
Abstract: The server device receives a model information from a plurality of terminal devices, and generates an integrated model by integrating the model information received from the plurality of terminal devices. The server device generates an updated model by learning a model defined by the model information received from the terminal device of update-target using the integrated model. Then, the server device transmits the model information of the updated model to the terminal device. Thereafter, the terminal device executes recognition processing using updated model.
-
公开(公告)号：US20210350522A1
公开(公告)日：2021-11-11
申请号：US17278361
申请日：2018-09-28
Applicant: NEC Corporation
Inventor： Asuka ISHII
Abstract: An information processing apparatus (2000) acquires observation information including a result of observing a structure (10) to which a moving load is applied, and determines, by using the observation information and a soundness condition, whether a relation between a temporal change in deflection amount of the structure (10) and a temporal change in an application position satisfies the soundness condition. The soundness condition is a condition, for deflection caused in the structure (10) by applying a load to the structure (10) while changing the application position, being satisfied when the structure (10) is sound. Then, the information processing apparatus (2000) outputs, based on a result of the determination, information relating to a degree of soundness of the structure (10).
-
公开(公告)号：US20210029297A1
公开(公告)日：2021-01-28
申请号：US17041092
申请日：2018-03-29
Applicant: NEC Corporation
Inventor： Gaku NAKANO , Kazuhito MURATA , Asuka ISHII , Masahiko OHTA
Abstract: A vibration measurement system comprises an image capturing apparatus, a distance measuring apparatus, a sensor that outputs a signal according to an inclination of the image capturing apparatus relative to the vertical direction, and a vibration measurement apparatus. The vibration measurement apparatus includes calculating an angle formed by the normal of an image capturing surface of the image capturing apparatus and the normal of the measurement target surface that the image capturing apparatus shoots, based on the signal output by the sensor, converting the image obtained that the image capturing apparatus shoots into an image that would be obtained were the normal of the measurement target surface coincident with the normal of the image capturing surface of the image capturing apparatus, using the calculated angle, and measuring vibration of the structure, using the converted image and the measured distance from the image capturing apparatus to the measurement target surface.
-
公开(公告)号：US20230334837A1
公开(公告)日：2023-10-19
申请号：US18026631
申请日：2020-09-24
Applicant: NEC Corporation
Inventor： Katsuhiko TAKAHASHI , Yuichi NAKATANI , Tetsuo INOSHITA , Asuka ISHII
IPC: G06V10/776 , G06V10/74
CPC classification number: G06V10/776 , G06V10/761
Abstract: In an object detection device, the plurality of object detection units output a score indicating a probability that a predetermined object exists for each partial region set with respect to inputted image data. The weight computation unit uses weight computation parameters to compute a weight for each of the plurality of object detection units on a basis of the image data and outputs of the plurality of object detection units, the weight being used when the scores outputted by the plurality of object detection units are merged. The merging unit merges the scores outputted by the plurality of object detection units for each partial region according to the weights computed by the weight computation unit. The first loss computation unit computes a difference between a ground truth label of the image data and the score merged by the merging unit as a first loss. Then, the first parameter correction unit corrects the weight computation parameters so as to reduce the first loss.
-
公开(公告)号：US20220277552A1
公开(公告)日：2022-09-01
申请号：US17624906
申请日：2019-07-11
Applicant: NEC Corporation
Inventor： Katsuhiko TAKAHASHI , Yuichi NAKATANI , Tetsuo INOSHITA , Asuka ISHII , Gaku NAKANO
IPC: G06V10/776 , G06V10/22 , G06V10/80
Abstract: In an object detection device, a plurality of object detection units output a score indicating the probability that a predetermined object exists for each partial region set with respect to inputted image data. On the basis of the image data, a weight computation unit uses weight computation parameters to compute weights for each of the plurality of object detection units, the weights being used when the scores outputted by the plurality of object detection units are merged. A merging unit merges the scores outputted by the plurality of object detection units for each partial region according to the weights computed by the weight computation unit. A loss computation unit computes a difference between a ground truth label of the image data and the scores merged by the merging unit as a loss. Then, a parameter correction unit corrects the weight computation parameters so as to reduce the computed loss.
-
Patent Agency Ranking
公开(公告)号：US20220113260A1
公开(公告)日：2022-04-14
申请号：US17426838
申请日：2019-02-01
Applicant: NEC Corporation
Inventor： Asuka ISHII
Abstract: An image processing apparatus includes a dividing unit, a measuring unit, a comparing unit, and a determining unit. The dividing unit spatially divides a time-series image of a structure surface captured during passage of a traffic load into a plurality of partial regions, and generates a plurality of partial time-series images. The measuring unit measures temporal changes in deflection amount of the structure surface in the respective partial regions from the plurality of partial time-series images. The comparing unit compares the temporal changes in deflection amount of the structure surface in the respective partial regions. The determining unit determines an orientation of the time-series image with respect to a passage direction of the traffic load based on a result of the comparison.
-
公开(公告)号：US20210033451A1
公开(公告)日：2021-02-04
申请号：US17042556
申请日：2018-03-29
Applicant: NEC Corporation
Inventor： Gaku NAKANO , Masahiko OHTA , Asuka ISHII , Kazuhito MURATA
Abstract: A vibration measurement apparatus 30 includes a detection unit 31 that acquires, as a pattern image from an image capturing apparatus 20 that shoots a measurement target surface of a structure 40, an image of the measurement target surface onto which pattern light is projected by an optical apparatus 10, and detects the pattern light from the pattern image, an estimation unit 32 that estimates an angle between the normal of the image capturing surface and the normal of the measurement target surface, based on the pattern light, an image conversion unit 33 that converts the shot image into an image that would be obtained were the normal of the measurement target surface coincident with the normal of the image capturing surface of the image capturing apparatus 20, using the estimated angle, and a vibration measurement unit 34 that measures the vibration of the structure 40 using the converted image.
-
公开(公告)号：US20240078832A1
公开(公告)日：2024-03-07
申请号：US18273930
申请日：2022-02-01
Applicant: NEC CORPORATION
Inventor： Asuka ISHII
CPC classification number: G06V40/107 , G06T7/73 , G06T2207/20084 , G06T2207/30196
Abstract: A learning-model generation apparatus 10 includes: an all-feature-amount-outputting unit that output, from image data of an object and for each joint of the object, a feature amount representing the joint; a feature-amount-generating unit that generates, from the feature amounts of the individual joints of the object and as training feature amounts, feature amounts in a case in which the feature amount of a certain joint is missing; and a learning-model-generating unit that, by using training data including the generated training feature amounts, generates a machine learning model by machine-learning positional relationships between the other joints in the case in which the feature amount of the certain joint is missing.
-
公开(公告)号：US20220398833A1
公开(公告)日：2022-12-15
申请号：US17775607
申请日：2019-11-13
Applicant: NEC Corporation
Inventor： Gaku NAKANO , Yuichi NAKATANI , Tetsuo INOSHITA , Katsuhiko TAKAHASHI , Asuka ISHII
IPC: G06V10/774 , G06N3/04
Abstract: The information processing device performs distillation learning of a student model using unknown data which a teacher model has not learned. The label distribution determination unit outputs an arbitrary label for the unknown data. The data generation unit outputs new generated data using an arbitrary label and unknown data as inputs. The distillation learning part performs distillation learning of the student model using the teacher model and using the generated data as an input.
-
公开(公告)号：US20220366678A1
公开(公告)日：2022-11-17
申请号：US17772622
申请日：2019-11-13
Applicant: NEC Corporation
Inventor： Asuka ISHII
IPC: G06V10/776 , G06V10/77 , G06V10/764 , G06V10/778 , G06T7/73
Abstract: Teacher and student models output inference results for training data. A loss calculation unit calculates a total loss using at least one of (1) a loss obtained by multiplying a difference between a true value and a student model output by a weight increasing as a confidence of the teacher model output is lower, (2) a loss obtained by multiplying a difference between the true value and the student model output by a weight increasing as a difference between the true value and the teacher model output is greater, and (3) a loss obtained by multiplying a difference between the teacher and student model outputs by weights increasing as the difference between the teacher and student model outputs is greater and increasing as the difference between the true value and the teacher model output is smaller. An update part updates parameters of the student model based on the total loss.
-
-
-
-
-
-
-
-
-
Search Results
11
records
(took 0.013 seconds)
