Is it a bird? Is it a plane? No, it's Apple's 3D recognition patent Apple secured a patent yesterday on software to create and identify 3D models of faces, animals, aircraft, military vehicles and tumours in one of the more unusual tech patents to be awarded in recent months. This came to light after the US Patent and Trademark Office published a series of newly awarded patents. Several other …
So kind of like:
http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=323810&tag=1
Does 18 years count as prior art? How come I can find prior art in 10 seconds and the USPTO cant find it (or their own arses) in the 3 or 4 hours it takes them to decide to award a patent.
"...provides useful geometric constraints for the analysis and extraction of man-made features in aerial imagery..."
That sounds like their recently-released maps product (and other companies' earlier-released maps products). Maybe the patent is for applying the technology to other uses, like eating peas with a spoon. If you use a spoon for that, you have to pay me!
I'd be moderately surprised to learn that this has no prior art... almost all the computer vision done today was invented in the 80s before anyone had enough processing power to make practical use of it. There's very little new under the sun.
Happily, it seems this patent is all about working with boring old simple 2D image sources, and who the hell cares about those anymore? All the cool kids are using 2-and-a-bit-D stereo imaging, range scans or proper 3D CAT scans these days.
One of my friends at uni did a 2d to 3d program in matlab for his final year project, it would create a 3d object of what was put infront of a webcam, rotating it would build a full 3d model of it. It didnt I.D. the objects, he probably should have done something with that, or would it have been the property of the uni :)
...never judge a patent by its title, but by the detailed contents. It's likely this is a specific implementation, presumably sufficiently different from the prior art to justify a patent (although if it's a US patent, you just never know).
You could easily get a patent for "A vehicle with 4 wheels" if it was a unique never-seen-before implementation and guarantee a ton of anti-patent coverage :)
This post has been deleted by its author
"I'm sorry sir but we can't buy the kit we need to identify the cause of your wife's life threatening condition as Apple have blocked it's sale in this country until the court case is finished."
OK - stretching a point I admit, but FFS - enough already with the patent crap.
I don't need any fancy techniques to identify the social tumor that is called Apple.
"I also have done this as part of a university computer vision course in 2006, and Apple gets a patent on it based on work done in 2010?"
Firstly, the patent dates from 2004. If you're going to argue there was stuff done before the date of the related patent application, it needs to pre-date 19 Aug 2004 (and as far as some complicated provisions of US patent law are concerned, it probably needs to date from 2003 or earlier).
Secondly, if you're arguing based on stuff done in 2003 or earlier, to be effective prior art it needs to have the following features:
A method for object recognition of a three dimensional (3D) object, the method comprising at an electronic device, maintaining a statistical 3D-shape model used to express 3D-shapes of object features in terms of a median 3D-shape, .mu., and a base of 3D-shapes, W, where each 3D-shape, t.sub.3D(i), is expressed in terms of an associated latent characteristic, u(i), as t.sub.3D(i)=W*u(i)+.mu., where 3D-shape vectors t.sub.3D and .mu. are d-dimensional, a matrix of the base of 3D-shapes is (d.times.q)-dimensional, and latent characteristic vectors u are q-dimensional, and where the object features include points, lines and contours corresponding to objects of an object class; accessing a profile corresponding to at least a known object of the object class, the profile stored on persistent storage communicatively coupled with the electronic device, the profile including a latent characteristic, u(k), associated with a 3D-shape, t.sub.3D(k), of the known object, in accordance with the statistical 3D-shape model, t.sub.3D(k)=W*u(k)+.mu.; detecting, by the electronic device, object features, t.sub.2D(o), of an object depicted in a two dimensional (2D) image, where a vector, t.sub.2D, representing representation of said detected object features is e-dimensional; determining, by the electronic device, a 3D-shape, t.sub.3D(o), corresponding to the detected object features, t.sub.2D(o), such that a projection model f(t.sub.3D) applied to the determined 3D-shape, t.sub.3D(o), results in the detected object features t.sub.2D(o)=f[t.sub.3D(o)], said determining the 3D-shape, t.sub.3D(o), comprising: optimizing the projection model, f[t.sub.3D(i)], applied to 3D-shapes, t.sub.3D(i), expressed in terms of respective latent characteristics, u(i), in accordance with the statistical 3D-shape model, t.sub.3D(i)=W*u(i)+.mu., said optimizing over the respective latent characteristics u(i), and selecting a latent characteristic u(o) corresponding to the optimized projection model, f[t.sub.3D(o)], such that the optimized projection model applied to a 3D shape expressed in terms of the selected latent characteristic, u(o), results in the detected object features t.sub.2D(o)=f[W*u(o)+.mu.)], where the 3D-shape expressed in terms of the selected latent characteristic, u(o), represents the determined 3D-shape, t.sub.3D(o); comparing, by the electronic device, the selected latent characteristic u(o), corresponding to the detected object features, t.sub.2D(o), at least with the latent characteristic included in the stored profile, u(k), and associated with the 3D-shape, t.sub.3D(k), of the known object; selectively recognizing the object depicted in the 2D image as the known object based on said comparing.
But please do continue to argue based on "what I reckon".
Anyone can express something in meaningless mathematical sounding terms, that in reality encompass a huge general swathe of stuff, that any lawyer can use to win a court case.
You would probably think it acceptable to patent --
"a device that In an x-y Cartesian coordinate system, with centre coordinates (a, b) and radius r is the set of all points (x, y) such that (x - a )^2 + ( y - b)^2 = r^2."
in other words - a circle.
"Face recognition may be an effective way of identifying a person without the cooperation or knowledge of the person."
Oh, great. So if a single photo identifying online you wasn't enough, a 3D model can be built and used as a match if your mug turns up on any camera or other image anywhere. Privacy? They've heard of it.
Time to start wearing my crash helmet all the time.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
