Title:
GESTURE POSE ESTIMATION METHOD BASED ON KALMAN FILTERING AND DEEP LEARNING
Document Type and Number:
WIPO Patent Application WO/2024/094227
Kind Code:
A1
Abstract:
The present invention relates to the technical field of computer vision, in particular to a gesture pose estimation method based on Kalman filtering and deep learning. The invention comprises two main parts: the first is the production of a pose estimation data set with high-precision pose annotation based on Kalman filtering data fusion, comprising a first-stage simulated hand pose estimation stage and a second-stage real gesture image acquisition and pose data fusion stage; and the second is performing 3D gesture pose estimation based on deep learning and integrated learning on dual-view RGB images, wherein the 3D gesture pose estimation is divided into a training stage and a prediction stage of a gesture pose estimation model. During the process of producing the data set, the gesture pose estimation method based on Kalman filtering and deep learning of the present application also needs to be used in observation correction in the Kalman filtering.
Inventors:
MA FENGYING (CN)
JI PENG (CN)
CAO MAOYONG (CN)
WANG XIANJIAN (CN)
ZHANG HUI (CN)
JI PENG (CN)
CAO MAOYONG (CN)
WANG XIANJIAN (CN)
ZHANG HUI (CN)
Application Number:
PCT/CN2023/139747
Publication Date:
May 10, 2024
Filing Date:
December 19, 2023
Export Citation:
Assignee:
QILU UNIV OF TECHNOLOGY SHANDONG ACADEMY OF SCIENCES (CN)
International Classes:
G06V40/10; G06V10/80; G06V10/82; G06V20/64
Foreign References:
| CN115410233A | 2022-11-29 | |||
| CN113408443A | 2021-09-17 | |||
| CN111860274A | 2020-10-30 | |||
| US20220262036A1 | 2022-08-18 |
Other References:
FENG JINXIANG 1657674261@QQ.COM; JI PENG JIPENG@QLU.EDU.CN; MA FENGYING MFY@QLU.EDU.CN: "Gesture Position Detection Based on Generative Adversarial Networks", 2022 13TH INTERNATIONAL CONFERENCE ON E-EDUCATION, E-BUSINESS, E-MANAGEMENT, AND E-LEARNING (IC4E), ACMPUB27, NEW YORK, NY, USA, 25 March 2022 (2022-03-25) - 27 March 2022 (2022-03-27), New York, NY, USA, pages 39 - 44, XP058854308, ISBN: 978-1-4503-9585-4, DOI: 10.1145/3529261.3529268
WANG XIANJIAN 10431200475@STU.QLU.EDU.CN; JI PENG JIPENG@QLU.EDU.CN; MA FENGYING MFY@QLU.EDU.CN: "End-to-end training of convolutional neural network for 3D hand pose estimation in dual-view RGB image", 2022 13TH INTERNATIONAL CONFERENCE ON E-EDUCATION, E-BUSINESS, E-MANAGEMENT, AND E-LEARNING (IC4E), ACMPUB27, NEW YORK, NY, USA, 25 March 2022 (2022-03-25) - 27 March 2022 (2022-03-27), New York, NY, USA, pages 45 - 49, XP058854307, ISBN: 978-1-4503-9585-4, DOI: 10.1145/3529261.3529269
WANG XIANJIAN 10431200475@STU.QLU.EDU.CN; JI PENG JIPENG@QLU.EDU.CN; MA FENGYING MFY@QLU.EDU.CN: "End-to-end training of convolutional neural network for 3D hand pose estimation in dual-view RGB image", 2022 13TH INTERNATIONAL CONFERENCE ON E-EDUCATION, E-BUSINESS, E-MANAGEMENT, AND E-LEARNING (IC4E), ACMPUB27, NEW YORK, NY, USA, 25 March 2022 (2022-03-25) - 27 March 2022 (2022-03-27), New York, NY, USA, pages 45 - 49, XP058854307, ISBN: 978-1-4503-9585-4, DOI: 10.1145/3529261.3529269
Attorney, Agent or Firm:
JINAN GEYUAN INTELLECTUAL PROPERTY AGENCY CO., LTD (CN)
Download PDF:
Previous Patent: DUAL-ENERGY X-RAY DETECTOR STRUCTURE AND DUAL-ENERGY X-RAY DETECTION METHOD
Next Patent: INDIRECT PATH FAILURE PROCEDURE IN MULTI-PATH
Next Patent: INDIRECT PATH FAILURE PROCEDURE IN MULTI-PATH