Smart Vision FAQs

What is Smart Vision? How’s Smart Vision different from the regular Cam Plus smart detection?

In the past, to build the AI to recognize an object on a webcam, you would first need to train an AI model. Think of it like teaching a child to learn. For almost all AI object detection tech on the market, the approach is to find the patterns among millions of training materials and build a model that fits most situations. This approach comes with a major disadvantage though: it cannot fit into every single personalized use case.

With Wyze Smart Vision, Wyze users are provided with an easy-to-use system that they can build, and train their own AI model for the specific use cases. A small example, you can train a model to detect if a door is opened or closed.

What is a Smart Vision skill?

Just like teaching a toddler to walk and talk, Smart Vision lets you teach Wyze cameras skills to detect specific things that you care about. You can teach it to detect if the stove is left on, or if a door is not closed and locked properly, for example. Once the skill is learned and equipped, the camera will be able to perform the skill and improve it over time.

Can I create multiple Smart Vision skills for one camera?

Yes, multiple skills can be create for each Smart Vision-equipped camera.

Is Smart Vision compatible with Cam Plus smart detection?

Smart Vision is different from the smart detection included with Cam Plus (Person, Package, Vehicle, Pet, etc.). If you train a Smart Vision model on your camera, you’ll only get Smart Vision detection results. If you would like to change back to Cam Plus smart detection, simply turn off the Smart Vision feature in the Wyze app.

Who has the access to Smart Vision?

Currently Smart Vision is available to Cam Plus users for testing. You’ll need an active Cam Plus license bound to the camera to be able to test Smart Vision.

What devices are supported by Smart Vision?

Smart Vision support is available on Wyze Cam v2. Support for more devices is coming later.

How do I set the area of interest?

While setting up a skill, you can set your Attention Zone by dragging the green box to where you want the camera to focus on. Make sure that you can tell the difference between your labels clearly, within the Attention Zone. This will help the camera learn the skill difference faster. 

Why can’t I change the Attention Zone once it has been set up?

Since Smart Vision learns the differences between different sets of images that are grouped under labels, changing the zone will essentially create too much noise and confuse the model. If the previous Attention Zone doesn’t work for you anymore, create a new skill.

Why is my skill not accurate? It's making mistakes!

The skill's accuracy depends on 3 things: data quality, data quantity, and how complex the job is. Training cameras to recognize apples from bananas with a clean contrast background can be fairly easy. But recognizing apples from peaches in a noisy, busy area can be much harder. 

When setting up a new skill, consider this: How clear is the difference between your labels? Is my Attention Zone set up properly to focus on the object? Does it exclude distractions and background noise? Are the images clear and good quality? Am I feeding each label enough data (images)?

While you're feeding good data to your camera to help it learn, our team is striving to improve the model so it performs better, more efficiently.

How long does it take to train a skill?

It depends on a few factors, and can take anywhere from seconds to hours. Training speed depends on how crowded the server is (how many other users are also training at the same time) and how many images are in your skill that need to be trained.

When can I re-train a skill?

You can re-train a skill at any time you want! If you find the model isn't accurate enough, feed it with more/better images to help it learn.

How do I set up notifications for my Smart Vision labels?

You can set up notifications while creating a skill.

Please note: In the current version, you can only create two labels under each skill, one being the default state and the other one being the of-interest state. For example, if you are setting up your skill to detect if you leave your door open, then door closed will be the default label, and door opened will be the of-interest label. Your Wyze Cam will send you notifications whenever it detects door opened.

Can I create more than two labels for one skill?

No, not at this time. You can currently only create two labels under each skill. Support for more labels is coming later - stay tuned!

Have more questions? Submit a request
Still need help?
Contact Us