Information processing method and device

Abstract

Information processing methods and devices are disclosed. The information processing method is applied to an information processing device. The information processing device contains a three-dimensional map pre-created for a specific spatial environment. A terminal device in the specific spatial environment can determine its own location based on the three-dimensional map. location, the method includes the following steps: obtaining an image taken by the terminal device; extracting feature points in the image, and obtaining feature descriptors used to characterize the feature points; based on the location of the terminal device and Depth information in the image is used to obtain the three-dimensional position of the feature point; the minimum spatial node in the specific spatial environment is determined; and the minimum spatial node to which each feature point belongs is determined based on the obtained three-dimensional position of each feature point. Points are clustered to obtain all feature descriptors contained in the minimum space nodes; and each minimum space node is managed based on a specific data structure.

Description

technical field [0001] The present invention relates to an information processing method and device, more specifically, to an information processing method and device capable of conveniently and quickly managing and updating a feature library. Background technique [0002] For an unknown environment, a 3D map can be pre-built by means of real-time localization and map construction (Simultaneous Localization and Mapping, SLAM) technology. In an environment where a map has been built, a mobile terminal device (such as a robot) obtains its own positioning information by matching the feature points in the image it captures with the feature points of the global map. In robot self-localization, since the scene of the robot is constantly changing, and the feature database of the global map has a large amount of data, it is very important to match the feature points in the image captured by the robot itself with the feature points in the feature database of the global map. time con...

AI Technical Summary

Problems solved by technology

In robot self-localization, since the scene of the robot is constantly changing, and the feature database of the global map has a large amount of data, it is very important to match the feature points in the image captured by the robot itself with the feature points in the feature database of the global map. time consuming

In addition, the operation of updating the feature library of the global map based on the images taken by the robot itself is also very time-consuming.

In addition, extending the feature library of the global map may involve complex data manipulation

Images