Processing

Please wait...

Settings

Settings

Goto Application

1. WO2021163103 - LIGHT-WEIGHT POSE ESTIMATION NETWORK WITH MULTI-SCALE HEATMAP FUSION

Publication Number WO/2021/163103
Publication Date 19.08.2021
International Application No. PCT/US2021/017341
International Filing Date 10.02.2021
IPC
G06T 7/70 2017.1
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
7Image analysis
70Determining position or orientation of objects or cameras
G06K 9/00 2006.1
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
CPC
G06K 9/6271
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
62Methods or arrangements for recognition using electronic means
6267Classification techniques
6268relating to the classification paradigm, e.g. parametric or non-parametric approaches
627based on distances between the pattern to be recognised and training or reference patterns
6271based on distances to prototypes
G06N 3/0454
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
04Architectures, e.g. interconnection topology
0454using a combination of multiple neural nets
G06N 3/084
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
08Learning methods
084Back-propagation
G06T 2207/10004
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
2207Indexing scheme for image analysis or image enhancement
10Image acquisition modality
10004Still image; Photographic image
G06T 2207/20084
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
2207Indexing scheme for image analysis or image enhancement
20Special algorithmic details
20084Artificial neural networks [ANN]
G06T 2207/30196
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
2207Indexing scheme for image analysis or image enhancement
30Subject of image; Context of image processing
30196Human being; Person
Applicants
  • NORTHEASTERN UNIVERSITY [US]/[US]
Inventors
  • FU, Yun
  • JIANG, Songyao
  • SUN, Bin
Agents
  • SOLOMON, Mark, B.
  • SMITH, James, M.
  • BROOK, David, E.
  • WAKIMURA, Mary Lou
  • CARROLL, Alice, O.
Priority Data
62/976,09913.02.2020US
Publication Language English (en)
Filing Language English (EN)
Designated States
Title
(EN) LIGHT-WEIGHT POSE ESTIMATION NETWORK WITH MULTI-SCALE HEATMAP FUSION
(FR) RÉSEAU D'ESTIMATION DE POSE LÉGER À FUSION MULTI-ÉCHELLE DE CARTE DE DENSITÉ
Abstract
(EN) Embodiments identify joints of a multi-limb body in an image. One such embodiment unifies depth of a plurality of multi-scale feature maps generated from an image of a multi-limb body to create a plurality of feature maps each having a same depth. In turn, for each of the plurality of feature maps having the same depth, an initial indication of one or more joints in the image is generated. The one or more joints are located at an interconnection of a limb to the multi-limb body or at an interconnection of a limb to another limb. To continue, a final indication of the one or more joints in the image is generated using each generated initial indication of the one or more joints.
(FR) Des modes de réalisation permettent d'identifier des articulations d'un corps à plusieurs membres dans une image. Un tel mode de réalisation unifie la profondeur d'une pluralité de cartes de caractéristiques multi-échelles générées à partir d'une image d'un corps à plusieurs membres pour créer une pluralité de cartes de caractéristiques ayant chacune une même profondeur. Puis, pour chaque carte de la pluralité de cartes de caractéristiques ayant la même profondeur, une indication initiale d'une ou plusieurs articulations dans l'image est générée. L'articulation ou les articulations sont situées au niveau d'une interconnexion d'un membre et du corps à plusieurs membres ou au niveau d'une interconnexion d'un membre et d'un autre membre. Ensuite, une indication finale de l'articulation ou des articulations dans l'image est générée à l'aide de chaque indication initiale générée de l'articulation ou des articulations.
Latest bibliographic data on file with the International Bureau