TechCrunch finally realizes that image recognition has not been solved
A new post at TechCrunch just appeared: Image Recognition Problem Finally Solved: Let’s Pay People To Tag Photos. A new company apparently provides image recognition for photo tagging – but only with human help! That’s not surprising to me. What is interesting is the change of attitude at TechCrunch: “A trail of failed startups have tried to tackle the problem… Google has effectively thrown in the towel…” After so many enthusiastic articles about image recognition technology somebody finally saw the light. And so did the founder of Riya. For a much longer list of “failed startups” in this area try this article about visual image search engines.
P.S. When I tried to reply to their post with a two-sentence comment, it was rejected. How odd!
P.S.S. The TechCrunch post was about TagCow, now I see a very recent post (elsewhere) about Picollator. They claim they have a visual image search engine for faces. People will keep trying….
When the fundamentals of vision are not even understood yet, and these folks are out there to make money out of it, failure is an obvious outcome. This however does not say anything about if there exists a solution to the computer vision problems, and if you ask me, I believe you can get there over a decade.
So, the enthusiasts should chill a bit, and restrict themselves to vision competitions like SRVC if they have something exciting to show.
Ajay, your 10 year prediction could be true but it’s also very pessimistic! In business terms 10 years means never. Consider this consequence: in the near future no company should even try to build a product if it would have to rely on image recognition.
Actually, an image search engine for faces exists already: in Google image, if you had an “imgtype=face” argument in the url the output is quite good.
You don’t need face recognition for that – all it takes is face _detection_. And yes, it’s pretty good.
Actually, Google is able to detect faces and it finds just portraits. Picollator engine searches for images by content and it enables to process and find group photos.
It sounds like Picollator is better than Google. That wouldn’t surprise me. But can it recognize faces?
Surely. It can recognize faces. It is able to find your photos if these photos are in picollator index.
Well, the only good way to verify your claim would be by adding two different images of the same person to the index and trying to find one with the other as a query. That’s not allowed.
You can see it now at http://www.picollator.com. Choose any photo from most popular list and click Find similar.
Do you know any engine which is able to find photos by pencil drawing? As for me I know just the sole engine:)