How Google is using AI to improve search, videos and more with a focus on visuals

Advertisement
How Google is using AI to improve search, videos and more with a focus on visuals
Unsplash
  • Google is adding more functionality to its search engine using AI.
  • It is also going to redesign the Google search with more visual results.
  • The improved Google search will allow users to search for topics visually.
Advertisement
Google explained how they are going to use artificial intelligence to make its search better during the recently concluded Search On livestream event. In a blog post, the company explained how its Multitask Unified Model works and how it can help you search better visually, learn more about topics you search and read related topics to videos you watch.

In this article, we will explain how each of these features will work and when it will be rolled out. Let’s dive in to find out more about how AI is going to make Google’s search better.

Searching Visually


Using MUM’s capabilities, Google is developing tools to search visually. As per the company, this will allow users to search in more natural and intuitive ways. This feature is expected to be rolled out in the coming months. Here’s how it will work.
Complimentary Tech Event
Transform talent with learning that works
Capability development is critical for businesses who want to push the envelope of innovation.Discover how business leaders are strategizing around building talent capabilities and empowering employee transformation.Know More

Whenever a user sees an image and wishes to search for a similar item, they can click on the Google lens icon and select the part of the image they wish to search about and type out their questions regarding it.

For example, If a user is looking at a picture of a shirt, and wishes to see the same pattern in socks then they just tap on the lens and select the pattern in the picture of the shirt and type out “Find same pattern socks”. It should deliver results with the same patterns.
Advertisement


This feature is likely to help users find something that might be difficult to describe just with words alone.

Redesign of Google Search


Google is adding “Things to Know” with each search result to help users understand what they are searching for. For example, if someone is searching the term “acrylic painting”, Google generally lists the top few hundred topics that are likely to help you in your quest. Along with it, it will also show you topics such as “how to make acrylic paintings with household items'' to help you understand more about the topic.

In the same redesign, it is also adding an option to broaden or refine a search topic. For example, if the user searched “acrylic painting”, they can refine their search and look for puddle pouring, or art classes that are associated with acrylic paint. The user can also broaden their search by checking out painting methods and famous painters.

Another addition to the Google search page is the addition of visual inspiration to the results page. This will help users easily browsable ideas from all across the web when they search for a topic. For example, if the user searches for “Halloween decorating ideas”, then the results would show a browsable image list from which the user can get inspiration.

Advertisement
This feature is rolled out and the users can try it on their systems.

Related topics to videos


Using AI, Google is already identifying key moments, but now they are also adding related topics to videos. When users are watching a video, they can explore more about the key elements in it by going through the related topics.

For example, if a user searches for “macaroni penguins” and views a video about them. Google will show related topics such as “macaroni penguin’s life story,” which may not be part of the search term or the video. This is done by understanding that topics contained in the video relate to this topic.

It can help users understand more about the topics they are researching. This feature will roll out in the coming weeks and Google is expected to add more visual enhancements in the future.

Earlier, the company had introduced how it was using AI to upscale low-resolution images. This was done by using two diffusion models to generate high fidelity images.
Advertisement
SEE ALSO:
How to use latitude and longitude in Google Maps to get the coordinates of a specific place or find a location
Google Meet is testing live translated captions, here’s how you can enable them
Google’s birthday: Here's a look at the significant milestones in Google's 23-year history
{{}}