Ask Your Question
-1

Multiple simultaneous logo/target recognition/reporting

asked 2013-10-25 05:40:11 -0600

updated 2013-10-25 06:14:01 -0600

berak gravatar image

Hi, we are planning an installation for our university grad show. The basic premise is that of a photobooth, with users placing vinyl logos/graphics on a glass screen. On the other side of the screen is a webcam.

We are looking for a way to have the webcam take a still image, recognise the logos present and email the captured image along with an array of detected logos (eg. [Image] + [ Apple, Android, Hat, Monacle, Banana].

From what we understand this should be well within the capabilities of any number of available AR frameworks. The trouble is we can't afford to tackle much of a learning curve in implementing this functionality.

What we are looking for is either: 1. Suggestions for a simple ready-made solution that requires only minor customisation 2. Expressions of interest from developers who can deliver the above-mentioned requirements

Thank you!

edit retag flag offensive close merge delete

1 answer

Sort by ยป oldest newest most voted
0

answered 2013-10-25 20:17:20 -0600

mccunnj gravatar image

Option A: train cascades for each logo. Then use object detection. Option B: use 2d features keypoints and descriptors, then find using matching (Flan or brute force)

Extraction and emailing is on you after that.

edit flag offensive delete link more

Question Tools

Stats

Asked: 2013-10-25 05:40:11 -0600

Seen: 459 times

Last updated: Oct 25 '13