Abstract: This document relates to performing live video stream analytics on edge devices. One example determines resources available to the system, and a video analytics configuration is selected that distributes work between edge devices and cloud devices in a cascading manner, where edge device processing is prioritized over cloud processing in order to conserve resources. This example can dynamically modify the allocation of processing depending on changing conditions, such as network availability.

Abstract: In accordance with implementations of the subject matter described herein, a new approach for creating a navigation trace is proposed. In these implementations, a reference signal is obtained, where the reference signal includes time series data collected by at least one environment sensor along a reference path from a start point to a destination. Then, a navigation trace is created for the reference path based on the obtained reference signal, where magnitudes at locations of the navigation trace reflect measurements collected by the at least one environment sensor at respective time points of the reference signal. In accordance with implementations of the subject matter described herein, a new approach for generating a navigation instruction from a navigation trace.

Abstract: In accordance with implementations of the subject matter described herein, a new approach for generating indoor navigation is proposed. Generally speaking, a reference signal that includes time series data collected by at least one environment sensor along a reference path from a start point to a destination is obtained. For example, the reference signal may be obtained by environment sensors equipped in a user's mobile device or another movable entity. Then, a movement event by identifying a pattern from the reference signal, the pattern describing measurements of the at least one environment sensor associated with a specific movement is extracted. Next, a navigation instruction is generated to indicate that the movement event occurs during a movement of the at least one environment sensor along the reference path. Further, the navigation instruction may be provided to a person for indoor navigation.

Abstract: In accordance with implementations of the subject matter described herein, a new approach for creating a navigation trace is proposed. In these implementations, a reference signal is obtained, where the reference signal includes time series data collected by at least one environment sensor along a reference path from a start point to a destination. Then, a navigation trace is created for the reference path based on the obtained reference signal, where magnitudes at locations of the navigation trace reflect measurements collected by the at least one environment sensor at respective time points of the reference signal. In accordance with implementations of the subject matter described herein, a new approach for generating a navigation instruction from a navigation trace.

Abstract: Although GPS has become a standard component of smartphones, providing accurate navigation during the last portion of a trip remains an important but unsolved problem. Despite extensive research on localization, the limited resolution of a map imposes restrictions on the navigation engine in both indoor and outdoor environments. To bridge the gap between the end position obtained from legacy navigation services and the real destination, a “last-mile” navigation system is proposed to enable plug-and-play navigation in indoor and semi-outdoor environments. The system exploits the ubiquitous, stable geomagnetic field and natural walking patterns to navigate the users to the same destination taken by an earlier traveler. Unlike existing localization and navigation systems, the proposed system is infrastructure-free, energy-efficient and cost-saving.

Abstract: Although GPS has become a standard component of smartphones, providing accurate navigation during the last portion of a trip remains an important but unsolved problem. Despite extensive research on localization, the limited resolution of a map imposes restrictions on the navigation engine in both indoor and outdoor environments. To bridge the gap between the end position obtained from legacy navigation services and the real destination, a “last-mile” navigation system is proposed to enable plug-and-play navigation in indoor and semi-outdoor environments. The system exploits the ubiquitous, stable geomagnetic field and natural walking patterns to navigate the users to the same destination taken by an earlier traveler. Unlike existing localization and navigation systems, the proposed system is infrastructure-free, energy-efficient and cost-saving.