Unsupervised Billboard Detection | Hacker News Reader

Unsupervised Billboard Detection

tech.adstruc.com

40 pointsmadhkrish12 comments

A great article. This is what Hacker News should be about; identifying a problem, researching solutions, and publishing the results for use. Thank you for a fantastic post.

Watch Google incorporate this into Street View to deliver billboard advertisements.

They actually filed a patent in 2008 to do that http://www.readwriteweb.com/archives/google_could_soon_augme... (direct link to the patent: http://appft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sec... )

This is great. I wonder if applying a crowdsource solution, with humans selecting the billboard, would also have delivered a cost-effective solution with the benefits in less time.

This is actually something we considered (we originally built this tool to import data from one of the major vendors). Their data was consistent enough that we were able to achieve a high enough hit rate, that it made sense to do it this way (OpenCV is an amazing framework for this type of task). However, from most of our vendors, the board themselves are not outlined/blocked out. In those cases, the CV algorithms fall apart fairly quickly, since quality of the input photos is quite variable (most are rather low resolution and you'd be surprised at the number of images we get with trees or other objects blocking a large portion of the board). The crowd sourced version is definitely on our fun tasks pile, since we've got a lot of inventory identify the bounds automatically.

Kind of surprised that the first comment isn't about automatically occluding these with google glass (yes, I know you can't fully paint over it http://blogs.valvesoftware.com/abrash/why-you-wont-see-hard-...)

Next up: a car, running this in real time, with an automated paintball launcher. Anybody got a good project name?

AdBlock Prius

I'm eagerly awaiting "They Live" mode.

I'd like to see how it works on the third image from the top, with the gas station sign, etc.

It is fairly straightforward to detect a billboard in a random nature scene; look for straight lines and intersections of them. Most (all?) billboards are also horizontal.

It is quite another thing to detect billboards in an urban setting.

I've added the output at the end of the blog post for that image... check it out!

This algorithm works pretty well for billboards in urban settings as well -- since the photos come with the units outlined.

Ah, this post makes me doubly excited about how I got OpenCV (+ the Ruby wrapper) to successfully install on Mountain Lion.

Reply on news.ycombinator.com