309 patents
Page 9 of 16
Utility
Method and System for Extraction and Annotation Using Semantic Attribute Paths
16 Dec 21
An annotation assistant for semi-automatic creation of annotated machine learning training data comprises a graphical annotation interface provider that provides a graphical annotation interface for displaying a sequence of tokens to be annotated and allowing a span selection by a user.
Salah AIT-MOKHTAR, Caroline BRUN, Agnes SANDOR
Filed: 19 Apr 21
Utility
Apparatus and Method for Dynamic Navigation of a Selected Geographical Zone
9 Dec 21
There is provided a handheld device for navigating about a selected geographical zone.
Sruthi VISWANATHAN, Cécile BOULARD, Maria Antonietta GRASSO, Fabien GUILLOT
Filed: 5 Jun 20
Utility
Adaptive Pointwise-pairwise Learning to Rank
9 Dec 21
A method of ranking items for a given entity uses sets of triplets <u, i, j>, each set of triplets including an entity u and a pair of items i and j with a known relative relevance for entity u, to train a learnable scoring function ƒ and to learn optimized values of a first set θ of learnable parameters.
Jean-Michel Renders, Yagmur Gizem Cinar
Filed: 29 Mar 21
Utility
Distillation of Part Experts for Whole-body Pose Estimation
2 Dec 21
A system for generating whole body poses includes: a body regression module configured to generate a first pose of a body of an animal in an input image by regressing from a stored body anchor pose; a face regression module configured to generate a second pose of a face of the animal in the input image by regressing from a stored face anchor pose; an extremity regression module configured to generate a third pose of an extremity of the animal in the input image by regressing from a stored extremity anchor pose; and a pose module configured to generate a whole body pose of the animal in the input image based on the first pose, the second pose, and the third pose.
Philippe WEINZAEPFEL, Romain BREGIER, Hadrien COMBALUZIER, Vincent LEROY, Gregory ROGEZ
Filed: 2 Jun 20
Utility
Method and System for Providing Translation for Conference Assistance
2 Dec 21
A method and a system for conference assistance provide, as a conference assistance tool, a screen including an original text that is a text input in one of a plurality of languages set to be used in a conference and a translation text that is translated from the original text in another language.
JoongHwi SHIN, Eunyoung LEE, Doosun YOO, Hoon CHOI, Min Sik PARK, Jongwon KIM, Bonghyun CHOI, Jeongmin KIM
Filed: 16 Sep 20
Utility
Method for training a convolutional recurrent neural network and for semantic segmentation of inputted video using the trained convolutional recurrent neural network
23 Nov 21
A method for training a convolutional recurrent neural network for semantic segmentation in videos, includes (a) training, using a set of semantically segmented training images, a first convolutional neural network; (b) training, using a set of semantically segmented training videos, a convolutional recurrent neural network, corresponding to the first convolutional neural network, wherein a convolutional layer has been replaced by a recurrent module having a hidden state.
Philippe Weinzaepfel
Filed: 22 Jul 19
Utility
Attention Over Common-Sense Network For Natural Language Inference
18 Nov 21
A training system includes: a first training dataset including first entries, wherein each of the first entries includes: a first sentence; a second sentence; and an indicator of a relationship between the first and second sentences; a training module configured to: generate a second dataset including second entries based on the first entries, respectively, wherein each of the second entries includes: the first sentence of one of the first entries; the second sentence of the one of the first entries; a first surface realization corresponding to first facts regarding the first sentence; the indicator of the one of the first entries; and train a model using the second dataset and store the model in memory.
Julien PEREZ
Filed: 14 May 20
Utility
Joint detection and description systems and methods
16 Nov 21
A system for detecting and describing keypoints in images is described.
Jérome Revaud, Cesar De Souza, Martin Humenberger, Philippe Weinzaepfel
Filed: 11 Dec 19
Utility
System and Method for Performing Cross-modal Information Retrieval Using a Neural Network Using Learned Rank Images
11 Nov 21
A system and method perform cross-modal information retrieval, by generating a graph representing the set of media objects.
Jean-Michel Renders, Stephane Clinchant, Thibault Formal
Filed: 8 Feb 21
Utility
Data placement and recovery for individually controlling storage device
9 Nov 21
Data placement and recovery technology for individually controlling a storage device includes a data management method that may achieve a power saving effect by distributing files between a portion of storage devices, for example, between storage devices included in a higher group and by limiting dependence according to a change in a state of the storage devices to be applied to a portion of storage devices to which a file distribution is performed.
Jaemyoun Lee, Chanyoung Park, Kyungtae Kang, Yun-cheol Choo, Sung-Won Jun, Taewoong Kim
Filed: 25 Jul 18
Utility
Adversarial Scene Adaptation for Camera Pose Regression
4 Nov 21
A pose estimation training system includes: a first model configured to generate a first 6 degrees of freedom (DoF) pose of a first camera that captured a first image from a first domain; a second model configured to generate a second 6 DoF pose of a second camera that captured a second image from a second domain, where the second domain is different than the first domain; a discriminator module configured to, based on first and second outputs from the first and second encoder modules, generate a discriminator output indicative of whether the first and second images are from the same domain; and a training control module configured to, based on the discriminator output, selectively adjust at least one weight value shared by the first model and the second model.
Boris CHIDLOVSKII
Filed: 29 Apr 20
Utility
Abstractive Multi-document Summarization Through Self-supervision and Control
4 Nov 21
A method for generating enriched training data for a multi-source transformer neural network for generation of a summary of one or more passages of input text comprises creating, from a plurality of input text sets, training points each comprising an input text subset of the input text set and a corresponding reference input text from the input text set, wherein the size of the input text subset is a predetermined number.
Matthias GALLE, Maximin COAVOUX, Hady ELSAHAR
Filed: 5 Mar 21
Utility
Method of Locating Points of Interest in a Geographic Area
28 Oct 21
There is provided a computer-implemented method of locating points of interest for a user in a geographic area.
Sruthi VISWANATHAN, Jean-Michel RENDERS, Bernard OMIDVAR-TEHRANI
Filed: 21 Jan 21
Utility
Method and system for subject-based ranking considering writer-reader interaction
26 Oct 21
A subject-based ranking determining method and system based on an interaction between a writer and a reader.
Jinhong Kim, Sanghoon Lee, Jin Hee Kim, Chan Hoon Park, Kwang Hyun Kim, Inho Kang
Filed: 3 Apr 18
Utility
Method for Computing an Itinerary from a Departure Location to an Arrival Location
21 Oct 21
A method creates a reduced set of feasible transfers T(t, L′) for a trip t of line L, for each target line L′ from a set of all transfers from line L to all other lines, by computing, for each origin line L, feasible transfers between stations of the origin line L and a destination line L′; sorting the computed feasible transfers to create a transfer set T(L); determining, for each trip t of origin line L, and for each transfer in the transfer set T(L), an earliest trip t′ of L′ wherein the transfer is feasible; and adding, for each trip t of origin line L, the determined transfer from t to t′ to the reduced set of feasible transfers T(t, L′) when trip t′ is the only destination trip of the transfers in the reduced set of feasible transfers T(t, L′) passing at the destination station and when it is earlier than all the previous destination trips of the transfers in the reduced set of feasible transfers T(t, L′) passing by the destination station.
Vassilissa Lehoux-Lebacque, Christelle Loiodice
Filed: 21 Apr 20
Utility
Method and system for sending bulk notifications
19 Oct 21
A bulk notification sending method includes, in response to sending a notification for providing content through a plurality of channels, retrieving subscribers of the channels connected to the content from a sending target database including information about subscribers for each channel; in response to a retrieved subscriber being a subscriber of a first channel among the channels, storing an identification value of the retrieved subscriber in each of a first storage for storing a sending target and a second storage for a redundancy removal and processing sending of a notification to the retrieved subscriber; and in response to the retrieved subscriber not being the subscriber of the first channel, retrieving the identification value of the retrieved subscriber from the second storage and determining whether to send the notification to the retrieved subscriber depending on whether the identification value of the retrieved subscriber is stored in the second storage.
Taekwan Kwon, Jungyoung Park, Seungmin Park
Filed: 20 Feb 20
Utility
Method and System for Providing Trending Search Terms
14 Oct 21
Methods and systems for providing trending search terms are provided.
Yuwon KIM, Gwang Lim GHIM
Filed: 7 Dec 20
Utility
End-To-End Graph Convolution Network
14 Oct 21
A natural language sentence includes a sequence of tokens.
Julien PEREZ, Morgan FUNTOWICZ
Filed: 12 Feb 21
Utility
Servers, non-transitory computer-readable media and methods for providing articles
12 Oct 21
A method of providing news articles by a server includes determining a plurality of clusters of articles among a plurality of articles, ranking each respective cluster among the plurality of clusters of articles based on issue information associated with the respective cluster to determine one or more highest ranked clusters, and causing an electronic device to display a graphical user interface including one or more representative articles corresponding to each of the one or more highest ranked clusters, each respective representative article among the one or more representative articles being selected from among two or more articles among the plurality of articles included in a corresponding cluster among the one or more highest ranked cluster.
BongSeak Yoo, Eunjoo Jung, Chang Yul Choi, JungHye Yun, Jungdal Jeon, Daewon Jeon, Jaemin Kim, Soo Hyang Kim, Hyejin Kim, Hyuntak Sung
Filed: 20 May 19
Utility
a Method for Training a Convolutional Neural Network for Image Recognition Using Image-conditioned Masked Language Modeling
7 Oct 21
A method and system pre-trains a convolutional neural network for image recognition based upon masked language modeling by inputting, to the convolutional neural network, an image; outputting, from the convolutional neural network, a visual embedding tensor of visual embedding vectors; tokenizing a caption to create a list of tokens, at least one token having visual correspondence to the image received by the convolutional neural network; randomly selecting one of the tokens in the list of tokens to be masked, the selected token being taken as ground truth; computing, using a language model neural network, hidden representations of the tokens; using the hidden representation of the masked token, as a query vector, to pool the visual embedding vectors in the visual embedding tensor, attentively; predicting the masked token by mapping the pooled visual embedding vectors to the tokens; determining a prediction loss associated with the masked token; and back-propagating the prediction loss to the convolutional neural network to tune parameters thereof.
Diane Larlus-Larrondo, Julien Perez, Mert Bulent Sariyildiz
Filed: 7 Apr 20