公开(公告)号：US11276184B2

公开(公告)日：2022-03-15

申请号：US16913437

申请日：2020-06-26

Applicant: FONDATION B-COM

Inventor： Albert Murienne ,  Laurent Launay

IPC: G06T7/246 ,  G06T7/50 ,  G06T7/73 ,  G06T17/00 ,  G06T19/20 ,  G06N20/00

Abstract: A method for determining the amplitude of a movement performed by a member of an articulated body comprises: obtaining a segment representative of the positioning of the member in a given reference frame at the end of said movement, generating a three-dimensional model of the member, positioned in said reference frame by means of the obtained segment, obtaining a cloud of three-dimensional points representing the member in said reference frame at the end of said movement, based on depth information provided by a sensor, said depth information defining a three-dimensional scene comprising at least a part of the articulated body including said member, repositioning the model of the member so as to minimize a predetermined error criterion between the obtained cloud of points and said model, and determining the amplitude of the movement, based on the new positioning of the model of the member.

公开(公告)号：US20220076438A1

公开(公告)日：2022-03-10

申请号：US17416104

申请日：2019-12-10

Applicant: FONDATION B-COM

Inventor： Nam-Duong Duong ,  Amine Kacete ,  Catherine Soladie

IPC: G06T7/55 ,  G06N20/20 ,  G06T7/80 ,  G06T19/00

Abstract: A method for predicting a three-dimensional representation in a three-dimensional scene, using one or more two-dimensional representations, obtained from at least one image generated by a camera. The method includes obtaining a forest of regression trees, and for the at least one image: extracting a 2D representation associated with one or more positions within the image; predicting a 3D representation, corresponding to the extracted 2D representation, using the regression tree forest includes a set of possible associations between at least one 2D representation and a 3D representation, each possible association resulting from a predictive model; evaluating one or more of the possible associations of the regression tree forest, according to a predetermined confidence criterion; and updating the regression tree forest including a deactivation of one or more possible associations, depending on the evaluation of the possible associations.

公开(公告)号：US20210407524A1

公开(公告)日：2021-12-30

申请号：US17362704

申请日：2021-06-29

Applicant: FONDATION B-COM

Inventor： Nicolas EPAIN ,  Francois SALMON

IPC: G10L19/008 ,  H04S7/00 ,  H04S3/02

Abstract: A conversion method comprises the following steps: for each of the signals (bE(t)) of a first set, determining values (α(f)) respectively associated with frequency bands; for each frequency band, converting the values (α(f)) associated with the relevant frequency band into at least one value representative of a virtual sound source oriented along the spatial direction associated with a data item stored for the relevant temporal frequency band; for each temporal frequency band, determining, on the basis of an above-mentioned representative value, a plurality of values (γ(f)) associated with the different signals (bS(t)) of the second set, respectively; constructing each signal (bS(t)) of the second set on the basis of the values (γ(f)) associated with this signal of the second set.

公开(公告)号：US20210263468A1

公开(公告)日：2021-08-26

申请号：US17255612

申请日：2019-05-28

Applicant: FONDATION B-COM

Inventor： Antonin GILLES ,  Patrick GIOIA

IPC: G03H1/08 ,  G02B27/01 ,  G03H1/00

Abstract: Disclosed is a method for generating a digital hologram representing a 3D scene including an object, an object being defined by points and their associated intensity. For each object, a prior step of calculating an “omnidirectional” angular spectrum of the light field emitted by an object in the scene on the surface of a geometric solid centered on the object, a surface of the solid being sampled according to a predetermined grid, a sample of the grid being associated with a vector frequency; for the scene, the following steps: —obtaining a pose of an observer in the world frame of reference; —deriving the hologram from the scene as a function of the pose obtained from the “multidirectional” angular spectra calculated for each object. The step of calculating an angular spectrum of the light field for each object of the scene takes into account predetermined viewing directions.

公开(公告)号：US20210058582A1

公开(公告)日：2021-02-25

申请号：US16961735

申请日：2019-07-26

Applicant: FONDATION B-COM

Inventor： Jean-Yves AUBIE ,  Wassim HAMIDOUCHE ,  Patrick DUMENIL ,  Glenn HERROU

IPC: H04N7/01

Abstract: A method for forming an image sequence that is an output sequence, from an input image sequence, is provided. The input image sequence has an input spatial resolution and an input temporal resolution. The output sequence has an output temporal resolution equal to the input temporal resolution and an output spatial resolution equal to a predetermined fraction 1/N of the input spatial resolution by an integer number N higher than or equal to 2. The method, implemented for a sub-sequence of the input frame sequence that is a current input sub-sequence and including a preset number of images, includes: obtaining a temporal frequency that is an image frequency, associated with the current input sub-sequence; processing the current input sub-sequence to obtain an output sub-sequence; and inserting the output sub-sequence and the associated image frequency into an output container.

公开(公告)号：US12022097B2

公开(公告)日：2024-06-25

申请号：US17427526

申请日：2020-01-24

Applicant: FONDATION B-COM

Inventor： Glenn Herrou ,  Wassim Hamidouche ,  Luce Morin ,  Jean-Yves Aubie

IPC: H04N19/33 ,  H04N19/105 ,  H04N19/119 ,  H04N19/132 ,  H04N19/139 ,  H04N19/147 ,  H04N19/176

CPC classification number: H04N19/33 ,  H04N19/105 ,  H04N19/119 ,  H04N19/132 ,  H04N19/139 ,  H04N19/147 ,  H04N19/176

Abstract: Disclosed is a method for encoding data corresponding to a video sequence, into a spatial resolution scalable binary flow, including a base layer and enhancement layer. The method includes, for a frame of the sequence: obtaining, from the frame, a first frame, partitioning the first frame into blocks, each block having a given size, and encoding a block of the first frame to generate the base layer; obtaining, from the frame, a second frame; partitioning the second frame into blocks by inferring the initial size of a block from the size of a corresponding block in the first frame, the block of the second frame having an initial spatial resolution; determining a spatial encoding resolution associated with the block from a set of predetermined spatial resolutions; and coding data representative of the at block, based on the determined spatial encoding resolution, to generate an enhancement layer.

公开(公告)号：US20240201313A1

公开(公告)日：2024-06-20

申请号：US18554936

申请日：2022-04-14

Applicant: FONDATION B-COM

Inventor： Luc LE MAGOAROU

IPC: G01S5/02

CPC classification number: G01S5/0284 ,  G01S5/0269

Abstract: A method for relative positioning of electronic apparatuses in communication with a communication device establishing a set of communication channels through which the device exchanges electromagnetic signals with each electronic apparatus includes, for each electronic apparatus, estimating a vector for which the elements represent, respectively, transmission through the communication channels between the device and the respective electronic apparatus, for at least first and second electronic apparatuses, determining a minimum distance from the distances existing, for all possible phase-shift values, between a first vector having a predetermined norm which is colinear to the vector estimated for the first electronic apparatus, and a second vector having the predetermined norm which is colinear to the vector estimated for the second electronic apparatus and modified by applying a phase shift having a value among the possible values, and determining a position indicator for the first electronic apparatus based on the determined distance.

公开(公告)号：US11943458B2

公开(公告)日：2024-03-26

申请号：US17848665

申请日：2022-06-24

Applicant: FONDATION B-COM

Inventor： Charles Bonnineau ,  Wassim Hamidouche ,  Jean-Yves Aubie

IPC: H04N19/30 ,  H04N19/13 ,  H04N19/132 ,  H04N19/186 ,  H04N19/42

CPC classification number: H04N19/30 ,  H04N19/13 ,  H04N19/132 ,  H04N19/186 ,  H04N19/42

Abstract: The invention relates to a method for decoding an image based on a base layer (Cbase) and enhancement information (Iam), said method comprising the following steps:



decoding said base layer (Cbase) in order to obtain at least one base component (Ydecbase, Udecbase, Vdecbase),
oversampling the at least one base component (Ydecbase, Udecbase, Vdecbase) in order to obtain at least one oversampled component (Ysurech, Usurech, Vsurech),
decoding the enhancement information (Iam) by means of a part (B) at least of artificial neural network in order to obtain enhancement values (Vam),
reconstructing the image based on the at least one oversampled component (Ysurech, Usurech, Vsurech) and the enhancement values (Vam).




The invention also relates to associated encoding method, electronic decoding device, electronic encoding device and signal.

公开(公告)号：US20230379506A1

公开(公告)日：2023-11-23

申请号：US18248023

申请日：2021-10-01

Applicant: FONDATION B-COM

Inventor： Félix HENRY ,  Gordon CLARE

IPC: H04N19/91 ,  H04N19/436

CPC classification number: H04N19/91 ,  H04N19/436

Abstract: A method for decoding a bit sequence includes applying, at the input of an artificial neural network, previously decoded values; as a result of the previously decoded values being applied, producing a context index at the output of the artificial neural network; and obtaining a new decoded value by applying a portion of the bit sequence to an entropic decoder parameterized in the context identified by the produced context index. An electronic decoding device and an associated computer program are also provided.

公开(公告)号：US20230188385A1

公开(公告)日：2023-06-15

申请号：US17921481

申请日：2021-04-14

Applicant: FONDATION B-COM

Inventor： Luc LE MAGOAROU ,  Stéphane PAQUELET

IPC: H04L25/02 ,  G06N3/08

CPC classification number: H04L25/0254 ,  G06N3/08

Abstract: Disclosed is a device for estimating a channel for communication system, which includes: a construction module designed to construct, on the basis of a physical model, a set of vectors) associated with a plurality of values of at least one parameter; an initialization module designed to initialize, as a function of the constructed vectors), columns of weighting coefficients defining at least a part of an artificial neural network; an application module designed to apply, as an input to the part of the artificial neural network, a vector determined as a function of noisy values so as to produce as an output a vector including estimated values; and a module for updating the weighting coefficients of the part of the artificial neural network by a learning technique. An associated method and computer program are also described.