Method and apparayus for depth-aided visual inertial odometry

• Main Authors: BAI DONG WOON, TYAGI ABHISHEK, LIANG YANGWEN, SETHURAMAN VIGNESH, LI XUE, WANG SHUANGQUAN
• Format: Patent
• Language: English; Korean
• Published: 01.12.2022
• Subjects: CALCULATING; COMPUTING; COUNTING; GYROSCOPIC INSTRUMENTS; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; MEASURING; MEASURING DISTANCES, LEVELS OR BEARINGS; NAVIGATION; PHOTOGRAMMETRY OR VIDEOGRAMMETRY; PHYSICS; SURVEYING; TESTING

The present invention is to provide a method and a user device for performing visual inertial odometry (VIO) in user equipment (UE). To this end, a method and an apparatus for depth-aided VIO are provided. Measurements are processed from an inertial measurement unit (IMU), a camera and a depth sensor of UE. Based on the processed measurements, keyframe residuals including at least depth residuals are determined. A sliding window graph is created and optimized based on factors derived from the keyframe residuals. An object pose of the UE is estimated based on the optimized sliding window graph. 시각적 관성 주행 거리 측정(VIO)을 수행하기 위한 방법 및 장치가 제공된다. 상기 UE의 관성 측정 유닛(IMU), 카메라 및 깊이 센서로부터 측정이 처리된다. 상기 처리된 측정에 기초하여 적어도 깊이 잔차를 포함하는 키프레임 잔차가 결정된다. 슬라이딩 윈도우 그래프가 상기 키프레임 잔차로부터 유도된 인자에 기초하여 생성 및 최적화된다. 상기 최적화된 슬라이딩 윈도우 그래프에 기초하여 상기 UE의 개체 포즈가 추정된다.