When you think of image search, typically, you think of going on to a search engine, such as Google, and type in a text query to get a series of images based on what you searched for. But have you ever thought about reverse image search? For those who isn’t sure what that is, reverse image is where you use an image as the search query, and the results will try bringing results to you, which should tell you what that image is. There have been several search engines who have been working on revers image search for years, including Google.
It seems that another engine has come into existence that allows for niche image searching, and it is called Wolfram Alpha. The new image search engine has launched theirs today at imageidentify.com.
The idea behind this image search identification engine is you place an image into the image search box, and it tries to show you what it thinks the image is. At this point, after playing around with several images I have, it makes me think that they’re still working on their overall database of image knowledge. The results were so far off, it was actually fun to see what the engine would come up with when I gave it an image. Luckily, at the bottom of the search results page, there was an input box that allows you to give them information on what the image actually is when it gets it wrong.
But honestly, I wasn’t surprised that there are these little mistakes, as doing a search the reverse way with images would be quite a bit more challenging.
You go to imageidentify.com and drag and drop the image on the home page:
Here’s what happens when you give the engine an image, and it gives you the correct information:
So what happens when you give it an image that it doesn’t recognize? This:
Overall, I think that this unique search engine could potentially work better as it grows, getting more feedback from users who are correcting it when the engine gets the results wrong.