Google's AR vice president explains what powers Lens

Google'southward AR vice president explains what powers Lens

Lens can identify over one billion products

Google Lens

In a recent Google Keyword blog mail service, Aparna Chennapragada the company's vice president of Google Lens and AR, shared what goes into making Lens piece of work.

Chennapragada says that she sees the camera equally the perfect tool for people to assemble information from since it's visual and humans are visual beings.

She then goes on to explicate that Lens utilizes a combination of machine learning, Google's Cognition Graph and computer vision to explain what it'south pointed at.

First off, Lens needs a fleck of training. Information technology does this by leveraging Google's vast library of online images to place objects.

Next, Lens uses Google's open source car learning framework TensorFlow to identify the images that the platform has decided match what yous're pointing it at.

The last slice of the puzzle is using the data that Lens has gained to match the item with Google'due south Noesis Graph. The Knowledge Graph is what builds the cards yous see at the pinnacle of Google Search results that summarize answers to questions or briefly explains something.

Lens takes data from this to requite its users the same handy facts that the Knowledge Graph does for spider web search.

After she explains Lens, Chennapragada shares that the Lens squad is working to aid the production place more items by using less than perfect photos of objects. To make this happen, Google is training Lens with photos taken from odd angles or in bad lighting, for example.

She also reveals that one of the reasons that Lens is so good at identifying text is because it uses the same spelling correction method every bit Google Search.

She also states that Lens tin now recognize over i billion products, a number that is four times the corporeality that it launched with last year.

Source: Google Keyword