Connected home camera manufacturers and service providers are looking for low cost machine learning solutions to process live video streams from cameras to detect people and objects in real-time and provide actionable alerts to their end users. Amazon Rekognition Streaming Video Events is a low cost, low latency, and fully managed machine learning service for detecting people, pets, and packages from live video streams. This service enables you to minimize false alerts from camera motion events by sending real time notifications only when a desired object such as a person, pet, or package is detected in the video frame.
Deliver relevant alerts in real time
Get accurate detection of people, pets, and packages even in varying lighting conditions, different camera angles, and camera resolutions. Choose the desired objects that you want to be notified on and deliver timely and actionable alerts to your users. Minimize false alerts from motion events by filtering out unnecessary notifications from wind, rain, shadows, or spiders and only focus on the objects that are relevant to you.
Amazon Rekogntion Streaming Video Events is built for connected home camera manufactures and service providers. Using the APIs, providers can reduce the time and cost of providing scalable object detection on video streams to users, while keeping up to date with the latest advancements in computer vision.
Improve connected home experiences
Create more elegant end-user experiences through intelligent home automation such as automatically turning on the garage light when a person is detected in the garage, creating a smart search feature in your companion app so your end users can find specific events of people, pets or packages, or integrating with voice assistants such as Alexa to enable Alexa announcements such as “package dropped off at the front door.”
How it works
Integrate with Amazon Kinesis Video Streams
Amazon Kinesis Video Streams (KVS) allows camera manufacturers and smart home service providers to easily stream live video from devices such as video doorbells, and indoor/outdoor cameras to AWS. Integrate your new or existing KVS streams with Amazon Rekognition Streaming Video Events to enable live video stream analysis.
Specify video duration
With Amazon Rekognition Streaming Video Events, you control how much video you need to process. You can specify the length of the video clips to be between 10 and 120 seconds. This provides you with the flexibility to better manage your machine learning inferencing costs.
Choose relevant objects
With Amazon Rekognition Streaming Video Events, you can choose one or more objects such as people, pets, and packages for detection from live video streams. In turn, this minimizes false alerts from camera motion events by sending notifications only when desired objects is detected in the video frame.
Create real-time alerts
As soon as Amazon Rekognition detects people, pets, or packages, it sends a smart alert that includes the video stream output with the detected label, bounding boxes, hero image, and time stamp. The service providers can send this smart alert to their user’s app or smart home device in real time.
Create bespoke experiences
The service providers can use Amazon Rekognition to create bespoke in-app experiences such as 'smart search' to find specific events of people, pets, or packages, ‘smart alerts’ with Alexa for announcements such as “A package was detected at the front door,” and more.
“We are always focused on making technology choices that could provide value to our customers and enable rapid growth while keeping the costs low. With Amazon Rekognition Streaming Video Events, we could launch person, pet and package detection at a fraction of the cost of developing everything ourselves. Every time a camera detects motion, we send alerts, in real-time, to our customers that are specific to events of interest such as people, pets or packages, and the latency is better than any other solution we’ve explored. This helps us filter out the noise and focus on what’s important to our customers–quality notifications. We have also improved our customer experience for video searches. Further, we could easily integrate this experience with Alexa and have not only doubled our customer engagement and satisfaction, but also seen a meaningful impact to our device sales. For us, it was obvious we didn’t want to build and maintain our own computer vision service. We turned to Amazon Rekognition’s APIs which are highly accurate, easy to incorporate into our systems, and very cost effective.”
Scott Beck, Chief Technology Officer - Abode Systems
“Simply relying on motion detection sensors triggers several alarms that are not a security or safety risk when there is a lot of activity in a scene. By utilizing machine learning to filter out the vast majority of such events, such as animals, shadows, moving vegetation and more, we can dramatically reduce the workload of the security operators and improve their efficiency.”
Ola Edman, Senior Director Global Video Development - 3xLOGIC
“With over 50,000 active cameras in the field, many without the advanced analytics of newer and more expensive camera models, 3xLOGIC takes on the challenge of false alarms every day. Building, training, testing, and maintaining computer vision models is resource-intensive and has a huge learning curve. With Amazon Rekognition Streaming Video Events, we simply call the API and surface the results to our users. It has been very easy to use and the accuracy is impressive.”
Charlie Erickson, Chief Technology Officer - 3xLOGIC