Visual search is quicker and easier than text based search
Have you ever spotted something and wondered what it was? You wanted to know immediately, but couldn’t accurately describe it using words in Google. It may have been a bug, a plant, or even an exotic fruit. With visual search technology, using pictures rather than text, that’s now possible. This is great for instances when you want an immediate answer on the go or don’t know exactly how to transcribe what you’re seeing into words.
As the technology powering mobile devices continues to improve a majority of people are using their phones as a computer. This means that less people are conducting searches on their desktop and are instead on the go looking for answers using their phone’s web browser or specific search apps such as Yelp or Craigslist. While this will get the job done, it certainly can be difficult for those who are short on time or individuals who find it difficult to type using a tiny smartphone keyboard.
As a result of this desire for instant answers and the change in behavior from switching from desktops to mobile devices, technology has evolved to meet our need for urgency and convenience. That’s where image recognition and visual search comes in.
There are two well known companies in image recognition space which provide this technology through a free mobile application, CamFind and Blippar. So which works best, which app should you download to try out visual search for yourself?
CamFind and Blippar had a battle in order to determine which technology was superior. Using the same image, the battle aimed to find which app could accurately identify none other than the Republican presidential candidate, Donald Trump. Check out this video to see what happened next.
We verified the results of the search demonstrated in this video and had similar findings.
CamFind successfully identifies Donald Trump and provides relevant links connecting users with polls about the politician, images related to “The Donald,” and the latest news surrounding the candidate. The search doesn’t stop there – additional results offer movies, videos, books, and even shopping results relevant to Donald Trump.
Blippar, on the other hand, presents general information relevant to the picture, but never identifies Donald Trump. The app then provides information such as definitions of search results, including “politician” or “administration”, and links to how to tie a tie.
We had a lot of fun with this battle and decided to conduct our own comparison of both apps using random objects in our office. Each time we searched the same conclusion seemed to hold true – CamFind worked more accurately, quickly, and with a greater ease of use. We could see that Blippar was trying to search for what we had presented to it, but in most cases was unable to accurately identify, or if it did, unable to provide valuable search links.
Visual search is a cool development in technology and it’s fun to use. It’s as easy as snapping a photo to be provided with endless information about objects in the physical world. As is clear from this battle, this technology is still evolving and has a long way to go before it is 100% perfected. This is due to the fact that the technology powering the search utilizes AI in order to continuously improve over time.
It will be interesting to see how this technology evolves and becomes more dominant in other devices such as wearables and connected homes. Until then, it is interesting to follow the developments of this technology in apps such as CamFind and Blippar.