Abstract: A method and device for extracting key frames in simultaneous localization and mapping and a smart device. The method includes acquiring an image frame from an image library storing a plurality of image frames of an unknown environment, and performing feature extraction on the image frame to obtain information of feature points, wherein the information includes a quantity of feature points; acquiring relative motion information of the image frame relative to the previous key frame, and calculating an adaptive threshold currently used by using the relative motion information; and selecting a key frame according to the information of feature points and the adaptive threshold indicating space information of image frames.

Abstract: A method and device for extracting key frames in simultaneous localization and mapping and a smart device. The method includes acquiring an image frame from an image library storing a plurality of image frames of an unknown environment, and performing feature extraction on the image frame to obtain information of feature points, wherein the information includes a quantity of feature points; acquiring relative motion information of the image frame relative to the previous key frame, and calculating an adaptive threshold currently used by using the relative motion information; and selecting a key frame according to the information of feature points and the adaptive threshold indicating space information of image frames.

Abstract: A binocular vision localization method, device and system are provided. The method includes calculating first pose change information according to two frames of images collected by a binocular camera unit at two consecutive moments and calculating second pose change information according to inertia parameters collected by an inertial measurement unit between the two consecutive moments. Matched feature points in the two frames are extracted from the two frames respectively. A reprojection error of each feature point is calculated. The calculations are taken as nodes or edges of a general graph optimization algorithm to acquire optimized third pose change information for localization. The system includes a binocular vision localization device, and a binocular camera unit and an inertial measurement unit respectively connected thereto, a left-eye camera and a right-eye camera are symmetrically located on two sides of the inertial measurement unit.

Abstract: A binocular vision localization method, device and system are provided. The method includes calculating first pose change information according to two frames of images collected by a binocular camera unit at two consecutive moments and calculating second pose change information according to inertia parameters collected by an inertial measurement unit between the two consecutive moments. Matched feature points in the two frames are extracted from the two frames respectively. A reprojection error of each feature point is calculated. The calculations are taken as nodes or edges of a general graph optimization algorithm to acquire optimized third pose change information for localization. The system includes a binocular vision localization device, and a binocular camera unit and an inertial measurement unit respectively connected thereto, a left-eye camera and a right-eye camera are symmetrically located on two sides of the inertial measurement unit.