METHOD FOR SEGMENTING SIGN LANGUAGE INTO MORPHEMES, METHOD FOR PREDICTING MORPHEME POSITIONS, AND METHOD FOR AUGMENTING DATA

Provided are a method for segmenting sign language into morphemes, a method for predicting morpheme positions, and a method for augmenting data. A system for analyzing sign language according to an embodiment of the present invention comprises: a recognition unit which recognizes key points of a spe...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: SHIN, Sa Im, PARK, Han Mu, JANG, Jin Yea
Format: Patent
Sprache:eng ; fre ; kor
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Provided are a method for segmenting sign language into morphemes, a method for predicting morpheme positions, and a method for augmenting data. A system for analyzing sign language according to an embodiment of the present invention comprises: a recognition unit which recognizes key points of a speaker from a sign language video; and a prediction unit which inputs the recognized key points into an artificial intelligence model, segments the sign language into morphemes, and predicts position information of the segmented morphemes. Accordingly, by recognizing the morphemes of the sign language video frame by frame on the basis of a skeletal model and thereby segmenting the sign language into morphemes and predicting morpheme positions, it is possible to lay the foundations for accurate sign language translation. La présente invention concerne un procédé de segmentation de langue des signes en morphèmes, un procédé de prédiction de positions de morphèmes, et un procédé d'augmentation de données. Un système d'analyse de langue des signes selon un mode de réalisation de la présente invention comprend : une unité de reconnaissance qui reconnaît des points clés d'un haut-parleur à partir d'une vidéo en langue des signes ; et une unité de prédiction qui entre les points clés reconnus dans un modèle d'intelligence artificielle, segmente la langue des signes en morphèmes, et prédit des informations de position des morphèmes segmentés. En conséquence, en reconnaissant les morphèmes de la vidéo en langue des signes trame par trame sur la base d'un modèle de squelette et en segmentant ainsi la langue des signes en morphèmes et en prédisant des positions de morphèmes, il est possible de poser les bases d'une traduction de langue des signes précise. 수어의 형태소 단위 분할 방법, 형태소 위치 예측 방법 및 데이터 증강 방법이 제공된다. 본 발명의 실시예에 따른 수어 분석 시스템은, 수어 동영상에서 화자의 키포인트들을 인식하는 인식부 및 인식된 키포인트들을 인공지능 모델에 입력하여, 수어를 형태소 단위로 분할하고 분할된 형태소의 위치 정보를 예측하는 예측부를 포함한다. 이에 의해, 골격 모델 기반으로 프레임 단위로 수어 동영상의 형태소를 인식하여, 수어를 형태소 단위로 분할하고 형태소 위치를 예측함으로써, 정확한 수어 번역의 토대를 마련할 수 있게 된다.