FIELD: data processing.
SUBSTANCE: invention relates to gesture recognition methods and systems. Forming a data array containing a gesture matrix and a gesture matrix associated with the gesture kireme matrix, including the kiremes and their configurations, for this purpose, gesture is gripped, divided into kiremes, from which kireme matrix is formed, each of the kireme is assigned a unique designation for the subsequent identification of the kireme in the kireme matrix and the gesture recognition in the gesture matrix associated with the kireme matrix, when dividing a gesture into kiremes, their sequence, number, distribution in time and in mutual temporal arrangement relative to each other in a gesture are taken into account and presented in the form of a relational formula for subsequent gesture generation by sending a gesture formula to the input of a graphic model of a person and reproducing the gesture in the gesture world by this graphic model. To recognize a gesture, a gesture image is captured, the gesture is divided into kiremes, which are compared with the kiremes in the kireme matrix and the kiremes are identified, at that, in the absence in the kireme matrix, the meaning of this kireme is specified and if the kireme is new, then it is included in the kireme matrix as a reference after the kirematic analysis and assignment of a unique designation to it, if the displayed kireme is a version of one of the kiremes existing in the matrix, then it is included in the kireme matrix as a version of the existing kireme with the assignment of a unique designation of the existing kireme, if the displayed kireme is recognized as an error in the display based on the results of the kirematic analysis of the gesture, this kireme is corrected in the gesture and repeated and captured for recognition.
EFFECT: high accuracy of recognizing a gesture of sign language, including taking into account the individual handwriting of the sign language.
7 cl, 8 dwg
Title | Year | Author | Number |
---|---|---|---|
ELECTRONIC SERVICE FOR STUDYING VOCABULARY AND SEMANTICS OF SIGN LANGUAGES | 2022 |
|
RU2807015C1 |
METHOD AND DEVICE FOR MARKING SIGN LANGUAGE GESTURES | 2023 |
|
RU2816047C1 |
GESTURE SHORTCUTS | 2010 |
|
RU2574830C2 |
STANDARD GESTURES | 2010 |
|
RU2534941C2 |
METHOD OF MULTIMODAL CONTACTLESS CONTROL OF MOBILE INFORMATION ROBOT | 2020 |
|
RU2737231C1 |
SYSTEM FOR RECOGNITION AND TRACKING OF FINGERS | 2012 |
|
RU2605370C2 |
METHOD OF CONTROLLING TELEVISION USING MULTIMODAL INTERFACE | 2010 |
|
RU2422878C1 |
METHOD OF CONTROLLING DEVICE USING GESTURES AND 3D SENSOR FOR REALISING SAID METHOD | 2011 |
|
RU2455676C2 |
METHOD AND DEVICE FOR PROCESSING VALUE DOCUMENTS | 2014 |
|
RU2667375C2 |
METHOD FOR GESTURE CONTROL AND THE DEVICE FOR ITS IMPLEMENTATION | 2021 |
|
RU2777910C1 |
Authors
Dates
2025-03-11—Published
2024-04-23—Filed