US 2026/0204100 A1 · JOINT SENSOR-CLOUD-BASED FALL DETECTION BASED ON LARGE VISION-LANGUAGE MODEL WITH MINIMAL FALSE ALARMS AND MISSED DETECTIONS

Abstract

In one aspect, a process of detecting personal falls is disclosed. This process may begin at a local vision sensor, which captures a video of a monitored space including a monitored person. A first fall detection model on the vision sensor is then used to process the captured video to detect a fall of the monitored person. Next, in response to detecting a fall of the monitored person, the local vision sensor transmits a raw image of the monitored space and a fall alert to a server. Next, the raw image of the monitored space and the associated fall alert is received by the server. A second fall detection model on the server is used to process the raw image to detect a fall of the monitored person. In response to determining that no fall has occurred of the monitored person, the server disregards the received fall alert, thereby reducing false alarms.

Claims22

Dependent claims

A computer-implemented method of automatically detecting personal falls through video monitoring on behalf of a user, the method comprising:

at a local vision sensor,
capturing a video of a monitored space including at least one monitored person;
processing one or more video frames of the video using a first fall detection model to detect a fall of the at least one monitored person;
and
in response to detecting a fall of the at least one monitored person, transmitting a raw image of the monitored space and a fall alert to a server;
and
at the server,
receiving the raw image of the monitored space and the associated fall alert from the local vision sensor;
processing the raw image using a second fall detection model to detect a fall of the at least one monitored person;
and
in response to determining that no fall has occurred of the at least one monitored person,
identifying the received fall alert from the local vision sensor as a false alarm;
and
disregarding the received fall alert, thereby reducing a number of false alarms.

11 more elements

A system for automatically detecting personal falls through video monitoring on behalf of a user, comprising:

a local vision sensor that includes:
one or more cameras configured to capture a video of a monitored space including at least one monitored person;
a first set of processors;
and
a first memory coupled to the one or more cameras and the first set of processors and storing instructions that, when executed by the first set of processors, cause the local vision sensor to:
process one or more video frames of the video using a first fall detection model to detect a fall of the at least one monitored person;
and
in response to detecting a fall of the at least one monitored person, transmit a raw image of the monitored space and a fall alert to a server;
and
the server that includes:
a second set of processors;
and
a second memory coupled to the second set of processors and storing instructions that, when executed by the second set of processors, cause the server to:
receive the raw image of the monitored space and the associated fall alert from the local vision sensor;
process the raw image using a second fall detection model to detect a fall of the at least one monitored person;
in response to determining that no fall has occurred of the at least one monitored person,
identify the received fall alert from the local vision sensor as a false alarm;
and
disregard the received fall alert, thereby reducing a number of false alarms;
and
in response to determining that a fall has occurred of the at least one monitored person, send a fall alert to a mobile device of the user indicating the at least one monitored person has fallen.

18 more elements

A computer-implemented method of detecting personal falls through video monitoring on behalf of a user, the method comprising:

at a local vision sensor,
capturing a video of a monitored space including at least one monitored person;
processing one or more video frames of the video using a first fall detection model to detect a fall of the at least one monitored person;
and
in response to determining that no fall has occurred of the at least one monitored person,
determining if a set of predefined conditions is met;
and
if so, transmitting a raw image of the monitored space and a non-fall decision to a server,
at the server,
receiving the raw image of the monitored space and the associated non-fall decision from the local vision sensor;
processing the raw image using a second fall detection model to detect a fall of the at least one monitored person;
and
in response to detecting a fall of the at least one monitored person by the second fall detection model:
identifying the received non-fall decision from the local vision sensor as a missed fall detection;
and
sending a fall alert to a mobile device of the user indicating the at least one monitored person has fallen, thereby mitigating the missed fall detection.

13 more elements

Dependencies are inferred from the claim text.

Joint sensor-cloud-based fall detection based on large vision-language model with minimal false alarms and missed detections

Life of this patent

Joint sensor-cloud-based fall detection based on large vision-language model with minimal false alarms and missed detections

Life of this patent

Abstract

Claims22