Ring cameras: IA descriptions instead of alerts

With this update of the cameras and ringing rings, there will necessarily be funny facts which will be recorded and broadcast on the networks. The fact remains that this update will now tell you what is happening in front of the camera of your front door or elsewhere.

As the name suggests, the new video description function generates textual notifications from the movements detected by the ringti and the Ring, Amazon ringtone.

From now on, instead of an alert indicating “gateway: person detected”, you will receive a message such as “a person placed a package and move away”. Or, again “a person presents himself with a black dog”.

These new descriptive alerts will appear in the notifications of the camera on your phone, allowing you to see at a glance if you should see the video recorded by the device.

For the moment in beta version and in English

The “video descriptions” function has just been launched in beta version for Ring Home Premium subscribers in the United States and Canada, but it is only offered in English. It will operate on all the ringtone and ring cameras currently available, according to Ring.

The “Video Descriptions” feature is added to that of “Smart Video Search” of the Ring cameras, launched last October, which makes it possible to question your cameras on recent events via the application, for example “have children left their bikes in the aisle? ». These two AI -based tools are available with the Ring Home Premium subscription, which also includes the 24 -hour and 7 days a week registration option.

This would “generate alerts only when an abnormal event occurs on your property”, explaining that Ring will be able to learn the routines of your home and send notifications only when something unusual happens. This means that you may not receive a alert if a person leaves the house with a broom, but you will be notified if your dog is putting the living room above.

Competition is also

Ring is not the only company to use AI to improve the notifications of cameras. Arlo and Google do the same bet: that of filtering alerts in the event of a false alarm, in particular thanks to facial recognition. Example, “your friend Pierre has just arrived by car in your entrance”.

Of course, these AI -based alert functions are in home automation (connected and intelligent houses) to filter and avoid fatigue of the many notifications.

And as in many things in terms of confidentiality, textual descriptions of activities could also facilitate the monitoring of people and be used potentially by malicious users.

For the short term, it will be necessary to discover the quality of video descriptions, if Ring knows how to make the difference between a person and a thief also passing.