APPARATUS AND METHOD FOR AUGMENTING DATA USING OBJECT SEGMENTATION AND BACKGROUND SYNTHESIS

Disclosed are an apparatus and method for augmenting data using object segmentation and background synthesis. According to the present invention, a large-scale image data set can be constructed by recognizing and extracting an object in an image rather than augmenting data through the simple positio...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	PARK YE SEONG, KIM HYEONG BOK, LEE JIN SUK, YOON SUK WON
Format:	Patent
Sprache:	eng ; kor
Schlagworte:	CALCULATING COMPUTING COUNTING IMAGE DATA PROCESSING OR GENERATION, IN GENERAL PHYSICS
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Beschreibung
Zusammenfassung:	Disclosed are an apparatus and method for augmenting data using object segmentation and background synthesis. According to the present invention, a large-scale image data set can be constructed by recognizing and extracting an object in an image rather than augmenting data through the simple position movement or rotation of an image and transforming the extracted object to generate various image data. 객체 분할과 배경 합성을 이용한 데이터 증강 장치 및 방법을 개시한다. 본 발명은 이미지의 단순한 위치 이동이나 회전 등을 통한 데이터 증강이 아니라 이미지 내의 객체를 인지하여 추출하고, 추출된 객체를 변환하여 다양한 이미지 데이터를 생성함으로써, 대규모 이미지 데이터 셋의 구축이 가능하다.