Tag Archive for: image search

Ever feel like words aren’t quite enough for what you want to ask Google? But, at the same time, the Google Image Search isn’t right for the job either? You’ll be excited to hear about Multisearch, the new way to use both text and images to find exactly what you’re looking for when searching the web.

Multisearch is a new feature in Google Lens, designed to deliver results based on contextual text phrases to better understand visual queries. In the announcement, Google says it designed the feature to “go beyond the search box and ask questions about what you see.”

How Multisearch Works

As part of Google Lens, Multisearch is still first-and-foremost about visual search. You start by opening the Google app on Android or iOS devices and uploading or taking a picture using your device’s camera. Then, you can provide more information about what you’re looking for by swiping up and tapping the “+ Add to your search” button.

Google offers a few examples of how people can use Multisearch to get better search results:

  • Screenshot a stylish orange dress and add the query “green” to find it in another color
  • Snap a photo of your dining set and add the query “coffee table” to find a matching table
  • Take a picture of your rosemary plant and add the query “care instructions”

In its current shape, Multisearch is best used for shopping search results. This means it is something e-commerce brands should definitely keep an eye on in the near future.

While the feature uses Google’s AI systems, the announcement clarifies it does not use the search engine’s most recent AI model, MUM – yet:

“We’re also exploring ways in which this feature might be enhanced by MUM– our latest AI model in Search– to improve results for all the questions you could imagine asking.”

Multisearch is available now to US users who have downloaded the most recent update of the Google app. For more information, check out the blog post revealing the feature here.

Bing LogoWhile Google may own a huge share of the search market, Bing is no slouch. They have undertaken a serious campaign to raise their profile and earn a larger amount of the searches happening every day. Part of this campaign is very straightforward; Bing wants to educate you about why they are a better search engine than Google.

To do this, Bing has undergone some high profile marketing (the “Bing it On” Challenge), but they’ve also upped their transparency and have been reaching out to interested consumers through the Bing Blog. Last week, Senior Program Manager Meenaz Merchant did just that, attempting to explain why Bing is the superior search engine for image searches.

Entity Understanding – Bing claims to be able to determine whether a search is for a person, place, or thing. By being able to distinguish the object of searches, they are able to deliver more accurate results to the user.

Big Data – According to Bing, Google doesn’t incorporate any image click data from the web or social signals into their search engine. Bing, on the other hand, includes user interaction data based on visual and text features to better match results for your interests.

Computer Vision Technologies – Bing uses deep learning to interpret dimensional data from the world to understand images more accurately than Google. They claim they are able to process images similar to the way the human brain does, allowing for more thorough cataloging and delivery of image results.

Thematic Intent Focus – Supposedly, Google treats image searches almost exactly how they handle text or web searches. Bing however says they are able to differentiate between the intent of users’ searches using specific methods for image searches.

Exact and Near Duplicate Images – Unlike Google, Bing is able to recognize and flag exact duplicates or near duplicate images from their searches, allowing for them to be filtered out of search results.

Aesthetics and Easy Viewing – Bing makes it easier to see multiple images in a single glance by more consistently sizing the results across the page. Google’s results tend to directly reflect their image size, creating a slightly jumbled view, while Bing’s is more cohesive, so that you can see all the images at once.

High Quality Images – Bing favors image quality above all, and they make it easy to see a full-size, high quality image after the first click. On other search engines, you often have to click through a page or two to actually see the full image.