Automatic video generation for navigation and object finding

摘要

Described is a system for conveying the spatial location of an object with respect to a user's current location utilizing a video rendering following an automatically generated path from the user's location to the location of the object of interest initiated from the user's current perspective. The system comprises a display device; a virtual three-dimensional model of the user's environment; a visualization creation module; a route planner; and a video rendering following an automatically generated path from the user's location to the location of the object of interest utilizing the visualization creation module, wherein the video rendering displayed on the display device is from a first-person view. Also described is a method of utilizing the system.

主权项

1. A system for conveying the spatial location of an object with respect to a user's current location, the system comprising:
a display device; at least one sensor for computing the user's current position and orientation relative to an object of interest; a virtual three-dimensional model of the user's environment generated from location information detected in six dimensions comprising x, y, z, roll, pitch, and yaw dimensions; a route planner configured to generate a route given a start location and an end location, wherein the start location is the user's current position and orientation and the end location is the location of the object of interest; and a visualization creation module configured to create a video rendering following the route created by the route planner by integrating the virtual three-dimensional model of the user's environment and an image of the user's current environment, wherein the video rendering begins with the image of the user's current environment and seamlessly transitions to a sequence of frames comprising the virtual three-dimensional model of the user's environment, wherein the video rendering is displayed on the display device starting from the user's current viewpoint as seen in the image of the user's current environment, and wherein the visualization creation module utilizes faster than real-time rendering of the sequences of frames for the video rendering.