-
After 15+ years, we've made a big change: Android Forums is now Early Bird Club. Learn more here.
SamuraiBigEd
Under paid Sasquatch!
Yep, I'm obviously not where I'm supposed to be
Are any of us really where we are supposed to be? Are we all just drifting through life ending up nowhere in particular or are we really at the place we need to be and just don't realize it?
I now return you to the regularly schedule thread...
andruoid
Android Expert
Are you sure this is the regular thread or is the regular thread in another time space ¿?
AngelArs
Well-Known Member
I have both. I prefer OS X BUT I also prefer that other people use windows so I can make more money
OK now I'm just getting dizzy...
AngelArs
Well-Known Member
And then again the resale value of a Mac is considerably higher also, so it all washes out
Don't forget about EFI-X
That's funny, I thought the base of this discussion was which platform was better. Mrspeedmaster shed some light on that and now you want to use a glass ceiling? I think mrspeedmaster did an excellent job of proving their point
That's funny, I thought the base of this discussion was which platform was better. Mrspeedmaster shed some light on that and now you want to use a glass ceiling? I think mrspeedmaster did an excellent job of proving their point
SamuraiBigEd's point was that we are talking about general consumer use. While Mrspeedmaster's example was valid for a very specific set of criteria, the application is impractical for the average consumer. I could easily cite specific examples of vertical market applications where a Mac couldn't even get their foot in the door ... POS systems or Enterprise ERP's off the top of my head, but that's hardly relevant to Joe and Jane Consumer looking for a home computer.
His example is equivalent to saying the Space Shuttle is a better vehicle because it actually works in space and can travel 17,000 mph but I wouldn't want to commute to work in it (unless, of course, I work in space.
)
AngelArs
Well-Known Member
Sorry but I respectfully disagree. The thread is "Mac VS PC what's better and why ?". It does NOT stipulate anywhere what is better for general consumers. You are assuming that. This is why I originally stated that the OP needed to be specific as to what is meant by 'better'. The term better will on its own be different to different people for many reasons, some of which won't have anything to do with the OS. If someone wants to learn how computers work then PC's are clearly better. If someone wants a virus free environment then Macs are better. I use both. I prefer Macs (many reasons). However I love seeing the hard core windows group, and I will never try to switch them into buying a Mac because truth be known windows users make me money. I can't make money on a Mac user unless they want to buy extra hardware. It's 6 of one and half dozen of another
The fact that Mrspeedmaster's was correct shows the foundation of OS X. Like Mrspeedmaster's I use whatever platform provides me with the tools I need.
I'm an equal opportunity hater. I find Windows, OS X and Linux (I run all three, each for its own job) to be equally lacking on all accounts.
I hope Android grows up a little more and just replaces all of them.
I totally mean this.
Until then, get the right OS for you, and if there no right OS for you, get the one that floats your boat.
And I'm sorry - with BSD unix under the hood, OS X is far from being inferior to FP toys - although that was hilarious.
I hope Android grows up a little more and just replaces all of them.
I totally mean this.
Until then, get the right OS for you, and if there no right OS for you, get the one that floats your boat.
And I'm sorry - with BSD unix under the hood, OS X is far from being inferior to FP toys - although that was hilarious.
SUroot
Extreme Android User
Yes indeed, with Parallels you dont even need to boot camp. Its not an OS thing for me though, as teh hardware is just so expensive to do that and legally you are obliged to pay for Windows ontop of that (unless apple sell MS licences with their devices?).
I only want Mac for one reason, to improve my support skills for our (work's) designer user base. I have a vmx of it but it wont boot in VMware player. I need to try and make one for myself.
So what if the Mac hardware seems expensive (to some)?
All of us are either on contract or paid full price to carry around what is effectively the hardware of an iPod Touch with celluar radio or two added on - one or two more chips and an antenna - for more than a doubling of the price of an iPod Touch. Sounds to me like counting coup on Mac expense is throwing stones in a glass house.
Like any consumer product, the price is just the price.
And price is not a reason to hate on it as containing a worthless OS.
OS X is a highly effective consumer port of NetBSD+FreeBSD+OpenBSD with a Mach microkernel on the bottom end and a nice consumer windowing system on the top end. When it first came out, my website (now defunct) showed how to run a Gnome desktop on it with all Linux apps (GNU apps actually, but most people think the GNU/Linux operating system is all just Linux) thanks to Fink, all at a hot key to switch between the two environments. (No virtual machine required.)
In addition to Parallels and VMWare, if interested in a good virtual machine without having to drop money -
https://www.virtualbox.org/
Universal virtual manager, good for owners of Mac, PC, Linux, Solaris.
It's important to run every operating system you can get because none of them get it right.
It's also important to just run the one that's right for you. And remember VirtualBox.
All of us are either on contract or paid full price to carry around what is effectively the hardware of an iPod Touch with celluar radio or two added on - one or two more chips and an antenna - for more than a doubling of the price of an iPod Touch. Sounds to me like counting coup on Mac expense is throwing stones in a glass house.
Like any consumer product, the price is just the price.
And price is not a reason to hate on it as containing a worthless OS.
OS X is a highly effective consumer port of NetBSD+FreeBSD+OpenBSD with a Mach microkernel on the bottom end and a nice consumer windowing system on the top end. When it first came out, my website (now defunct) showed how to run a Gnome desktop on it with all Linux apps (GNU apps actually, but most people think the GNU/Linux operating system is all just Linux) thanks to Fink, all at a hot key to switch between the two environments. (No virtual machine required.)
In addition to Parallels and VMWare, if interested in a good virtual machine without having to drop money -
https://www.virtualbox.org/
Universal virtual manager, good for owners of Mac, PC, Linux, Solaris.
It's important to run every operating system you can get because none of them get it right.
It's also important to just run the one that's right for you. And remember VirtualBox.
SUroot
Extreme Android User
Woah there Nellie, hold your horses!
I dont hate on the OS because of the hardware prices. I think I wasn't clear in my post.
These things accumulate my "Hate" of Apple in general.
It all started when the iPod was released and it was seen as the best invention in recent years. Mainly that's society and their blindness coupled with Apple and their marketing "genius".
They claim to invent, create etc when there had been products on the market like that already. Tablets had been around for years. "facetime" already existed. Touch screens already existed. MP3 players already existed.
They rarely invent. I admit that they do often market something in such a way that people who had never been interested in this technology - suddenly are and that helps to revive certain "flopped" technology (tablets for example). I admit also from time to time they have imporved on existing ideas.
I don't hate apple products because I hate apple as a company. I hate/dislike apple products for their own individual reasons. This is then exacerbated by the company I hate for other reasons.
I Dislike the ipod for these reasons:
If you add to this that apple (inadvertantly) "culture" their user base to argue/justify their product is the best, simply because its apple without any technical basis or knowledge of alternative products and their capabilities.
I actually makes me boil under the skin. I unfortunately use this (not quite blind hatred) to pre-judge people. I see someone with an apple product, I cant help thinking their an idiot. I know this isnt the case (not always anyway) but I can't help it.
In short, I feel just in blaming apple not only for forcing upon me an attitude of "I cant take any of your advice seriously, you've got an iSheep" but also for destroying what little faith in humanity and man kind I had left.
EDIT> In regards to virtual machines, VMware doesn't work on Mac but there is no need as parallels is good enough. As for PC / Linux, VMware player is Free for non-commerical use and is much more stable than Oracle's Virtual Box.
I'm actually looking at licences for both of these at work (Vmware Workstation and Virtualbox).
Where as Virtual box is
I dont hate on the OS because of the hardware prices. I think I wasn't clear in my post.
- I hate the OS because I think the OS is crap (mainly restrictions - I had already mentioned this earlier, but I didnt reference this in my last post Apologies).
- I hate the hardware as I think it is over priced
- I hate Apple pricing as the cost includes the cost of a proprietory OS based on another, existing OS
- I hate Apple as it's their way or the highway (restrictions again)
These things accumulate my "Hate" of Apple in general.
It all started when the iPod was released and it was seen as the best invention in recent years. Mainly that's society and their blindness coupled with Apple and their marketing "genius".
They claim to invent, create etc when there had been products on the market like that already. Tablets had been around for years. "facetime" already existed. Touch screens already existed. MP3 players already existed.
They rarely invent. I admit that they do often market something in such a way that people who had never been interested in this technology - suddenly are and that helps to revive certain "flopped" technology (tablets for example). I admit also from time to time they have imporved on existing ideas.
I don't hate apple products because I hate apple as a company. I hate/dislike apple products for their own individual reasons. This is then exacerbated by the company I hate for other reasons.
I Dislike the ipod for these reasons:
- It is not and never has been the best MP3 player on the market.
- It is one of the most expensive MP3 players on the market.
- Its made by apple who (although its their prerogative) think its reasonable to charge the earth for a sub-par device
- Everyone has one because they are oblivious to the above or simply dont care
If you add to this that apple (inadvertantly) "culture" their user base to argue/justify their product is the best, simply because its apple without any technical basis or knowledge of alternative products and their capabilities.
I actually makes me boil under the skin. I unfortunately use this (not quite blind hatred) to pre-judge people. I see someone with an apple product, I cant help thinking their an idiot. I know this isnt the case (not always anyway) but I can't help it.
In short, I feel just in blaming apple not only for forcing upon me an attitude of "I cant take any of your advice seriously, you've got an iSheep" but also for destroying what little faith in humanity and man kind I had left.
EDIT> In regards to virtual machines, VMware doesn't work on Mac but there is no need as parallels is good enough. As for PC / Linux, VMware player is Free for non-commerical use and is much more stable than Oracle's Virtual Box.
I'm actually looking at licences for both of these at work (Vmware Workstation and Virtualbox).
Where as Virtual box is
Yeah, ok, fair enough.
However my post wasn't directed at you exclusively, it simply contained elements sparked by reading a number of posts and was just directed at providing some info along with opinions.
As for the OS being crap, if I followed your argument, that's based on it being unsuitable for IT use. Agreed, Windows seems much better for that for most organizations.
But you listed Linux as your first OS choice. Linux is a kernel by Linus and unix apps from the GNU repository, as in GNU's Not Unix (recursive name). One of the first open source projects directed at bringing commercial unix out from under corporate control. Along the same lines, unix was brought much further along by the students at Berkeley, and became the Berkeley Systems Distribution. When Berkeley closed the BSD Lab and kicked it out of the nest, the system forked. OpenBSD (source of any open security you use today, like ssh), NetBSD (unix networking, as stolen by Microsoft to finally get TCP/IP to work) and FreeBSD (emphasis on portability). Those were fully elaborated unix distributions when GNU/Linux was in its infancy.
OS X is a fusion of those distributions along with some eye candy and apps. Hardly crap, but that's just my opinion as a long-time unix user.
Maybe I don't get out enough or maybe I'm just an equal opportunity hater. I've met Apple users who think they rule the planet, Windows users who think they've accomplished something because they've learned how to maintain a system with antiquated constructs and poorly implemented by Microsoft, and Linux users who seem to have never left mom's basement. Yeah, ok, I hate all of those, they're all of the same type. All attitude and no information.
The internet as you know it exists because of BSD. Not just a little - whole cloth.
I simply don't agree that the modern personal desktop version of the operating system that spawned the internet is crap.
PS - As for VMWare not working on a Mac, that's not entirely true. I have it installed on one of my lab Mac machines and last I checked, it was working fine.
http://www.vmware.com/products/fusion/overview.html
For the home user, I still recommend VirtualBox. Most people seem to just want to run Linux completely for free and it serves that purpose very, very well. Again, you can run Linux under a virtual machine on a Mac - or just build your own desktop from open sources, run OS X's native X11, and unix away to your heart's content.
However my post wasn't directed at you exclusively, it simply contained elements sparked by reading a number of posts and was just directed at providing some info along with opinions.
As for the OS being crap, if I followed your argument, that's based on it being unsuitable for IT use. Agreed, Windows seems much better for that for most organizations.
But you listed Linux as your first OS choice. Linux is a kernel by Linus and unix apps from the GNU repository, as in GNU's Not Unix (recursive name). One of the first open source projects directed at bringing commercial unix out from under corporate control. Along the same lines, unix was brought much further along by the students at Berkeley, and became the Berkeley Systems Distribution. When Berkeley closed the BSD Lab and kicked it out of the nest, the system forked. OpenBSD (source of any open security you use today, like ssh), NetBSD (unix networking, as stolen by Microsoft to finally get TCP/IP to work) and FreeBSD (emphasis on portability). Those were fully elaborated unix distributions when GNU/Linux was in its infancy.
OS X is a fusion of those distributions along with some eye candy and apps. Hardly crap, but that's just my opinion as a long-time unix user.
Maybe I don't get out enough or maybe I'm just an equal opportunity hater. I've met Apple users who think they rule the planet, Windows users who think they've accomplished something because they've learned how to maintain a system with antiquated constructs and poorly implemented by Microsoft, and Linux users who seem to have never left mom's basement. Yeah, ok, I hate all of those, they're all of the same type. All attitude and no information.
The internet as you know it exists because of BSD. Not just a little - whole cloth.
I simply don't agree that the modern personal desktop version of the operating system that spawned the internet is crap.
PS - As for VMWare not working on a Mac, that's not entirely true. I have it installed on one of my lab Mac machines and last I checked, it was working fine.
http://www.vmware.com/products/fusion/overview.html
For the home user, I still recommend VirtualBox. Most people seem to just want to run Linux completely for free and it serves that purpose very, very well. Again, you can run Linux under a virtual machine on a Mac - or just build your own desktop from open sources, run OS X's native X11, and unix away to your heart's content.
SUroot
Extreme Android User
I like linux as its fairly new to me and Im geeky. I like getting right down into it.
I still recommend VMware player for the home linux user / Windows user. Player itself doesnt work on mac but I wasnt aware of a different vmware product for mac. Thanks for the info.
I still recommend VMware player for the home linux user / Windows user. Player itself doesnt work on mac but I wasnt aware of a different vmware product for mac. Thanks for the info.
SamuraiBigEd
Under paid Sasquatch!
I also have no problem with the computer (now that they run Intel processors) or OS itself, just the company and it's business practices, locked down ecosystem and patent-whoring policies.
While a subject for another thread as example I present Apple filing for a patent on Face Recognition on a smartphone!
On edit: The media reacts like this is something totally Apple.
Apple To Make Passwords Obsolete With New Face Recognition Patent - Forbes
Second edit: OK, I am forced to do a little back tracking on this one. After posting I went to the USPTO website and searched the application published on 12/29/2011 and it is another example of government inefficiency at work, the application was filed 6/29/2010 which could mean Android makers will once again be out of luck in a patent suit.
For those interested:
While a subject for another thread as example I present Apple filing for a patent on Face Recognition on a smartphone!
On edit: The media reacts like this is something totally Apple.
Apple To Make Passwords Obsolete With New Face Recognition Patent - Forbes
Second edit: OK, I am forced to do a little back tracking on this one. After posting I went to the USPTO website and searched the application published on 12/29/2011 and it is another example of government inefficiency at work, the application was filed 6/29/2010 which could mean Android makers will once again be out of luck in a patent suit.
For those interested:
( 1 of 512 )
United States Patent Application 20110317872
Kind Code A1
Free; Robert Mikio December 29, 2011
Low Threshold Face Recognition
Abstract
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, are disclosed for reducing the impact of lighting conditions and biometric distortions, while providing a low-computation solution for reasonably effective (low threshold) face recognition. In one aspect, the methods include processing a captured image of a face of a user seeking to access a resource by conforming a subset of the captured face image to a reference model. The reference model corresponds to a high information portion of human faces. The methods further include comparing the processed captured image to at least one target profile corresponding to a user associated with the resource, and selectively recognizing the user seeking access to the resource based on a result of said comparing.
Inventors: Free; Robert Mikio; (San Jose, CA)
Assignee: APPLE INC.
Cupertino
CA
Serial No.: 826581
Series Code: 12
Filed: June 29, 2010
Current U.S. Class: 382/103; 382/118
Class at Publication: 382/103; 382/118
International Class: G06K 9/00 20060101 G06K009/00
Claims
1. A method performed by an image processor, the method comprising: processing a captured image of a face of a user seeking to access a resource by conforming a subset of the captured face image to a reference model, the reference model corresponding to a high information portion of human faces; comparing the processed captured image to at least one target profile corresponding to a user associated with the resource; and selectively recognizing the user seeking access to the resource based on a result of said comparing.
2. The method of claim 1, wherein the high information portion includes eyes and a mouth.
3. The method of claim 2, wherein the high information portion further includes a tip of a nose.
4. The method of claim 2, wherein said processing the captured image comprises detecting a face within the captured image by identifying the eyes in an upper one third of the captured image and the mouth in the lower third of the captured image.
5. The method of claim 4, wherein the reference model includes a reference image of a face, and wherein said processing the captured image further comprises matching the eyes of the detected face with eyes of the face in the reference image to obtain a normalized image of the detected face.
6. The method of claim 5, wherein said processing the captured image further comprises vertically scaling a distance between an eyes-line and the mouth of the detected face to equal a corresponding distance for the face in the reference image in order to obtain the normalized image of the detected face.
7. The method of claim 6, wherein said processing the captured image further comprises matching the mouth of the detected face to the mouth of the face in the reference image in order to obtain the normalized image of the detected face.
8. The method of claim 5, wherein said comparing the processed captured image comprises: obtaining a difference image of the detected face by subtracting the normalized image of the detected face from a normalized image of a target face associated with a target profile; and calculating scores of respective pixels of the difference image based on a weight defined according to proximity of the respective pixels to high information portions of the human faces.
9. The method of claim 8, where the weight decreases with a distance from the high information portions of the human faces.
10. The method of claim 9, wherein the weight decreases continuously with the distance from the high information portions of the human faces.
11. The method of claim 9, wherein the weight decreases discretely with the distance from the high information portions of the human faces.
12. The method of claim 9, wherein the weight decreases from a maximum weight value at a mouth-level to a minimum value at an eyes-line.
13. The method of claim 1, wherein said selectively recognizing the user comprises presenting to the user a predetermined indication according to a user's profile.
14. The method of claim 1, wherein the resource represents an appliance, the method further comprising capturing the image using an image capture device of the appliance.
15. The method of claim 14, wherein said selectively recognizing the user comprises turning on a display of the appliance, wherein the display had been off prior to said comparing.
16. The method of claim 14, wherein said processing the captured image comprises: applying an orange-distance filter to the captured image; and segmenting a skin-tone orange portion of the orange-distance filtered image to represent a likely presence of a face in front of the image capture device.
17. The method of claim 16, wherein said processing the captured image further comprises determining changes in area and in location of the skin-tone orange portion of the captured image relative to a previously captured image to represent likely movement of the face in front of the image capture device.
18. The method of claim 17, wherein said processing the captured image further comprises detecting a face within the skin-tone orange portion of the orange-distance filtered image when the determined changes are less than predetermined respective variations.
19. An appliance comprising: a data storage device configured to store profiles of users associated with the appliance; an image capture device configured to acquire color frames; one or more data processors configured to perform operations including: apply an orange-distance filter to a frame acquired by the image capture device; determine respective changes in area and location of a skin-tone orange portion of the acquired frame relative to a previously acquired frame; infer, based on the determined changes, a presence of a face substantially at rest when the frame was acquired; detect a face corresponding to the skin-tone orange portion of the acquired frame in response to the inference, the detection including finding eyes and a mouth within the skin-tone orange portion; normalize the detected face based on locations of eyes and a mouth of a face in a reference image; analyze weighted differences between normalized target faces and the normalized detected face, the analysis including weighting portions of a face based on information content corresponding to the portions, and the target faces being associated with respective users of the appliance; match the face detected in the acquired frame with one of the target faces based on a result of the analysis; and acknowledge the match of the detected face in accordance with a profile stored on the data storage device and associated with the matched user of the appliance.
20. The appliance of claim 19, wherein the data storage device is further configured to store: rules for analyzing the weighted differences including weighting rules and scoring rules; and rules for matching the detected face against target faces.
Description
BACKGROUND
[0001] This specification relates to low threshold face recognition, e.g., a face recognition system that can tolerate a certain level of false positives in making face recognition determinations.
[0002] Most face recognition systems fall into one of two categories. A first category system tends to be robust and can tackle various lighting conditions, orientations, scale and the like, and tends to be computationally expensive. A second category system is specialized for security-type applications and can work under controlled lighting conditions. Adopting the first category systems for face recognition on consumer operated portable appliances that are equipped with a camera would unnecessarily use an appliance's computing resources and drain its power. Moreover, as the consumer portable appliances tend to be used both indoor and outdoor, the second category systems for face recognition may be ineffective. Such ineffectiveness may be further exacerbated by the proximity of the user to the camera, i.e., small changes in distance to and tilt of the appliance's camera dramatically distort features, making traditional biometrics used in security-type face recognition ineffective.
SUMMARY
[0003] This specification describes technologies relating to reducing the impact of lighting conditions and biometric distortions, while providing a low-computation solution for reasonably effective (low threshold) face recognition that can be implemented on camera-equipped consumer portable appliances.
[0004] In general, one aspect of the subject matter described in this specification can be implemented in methods performed by an image processor that include the actions of processing a captured image of a face of a user seeking to access a resource by conforming a subset of the captured face image to a reference model. The reference model corresponds to a high information portion of human faces. The methods further include comparing the processed captured image to at least one target profile corresponding to a user associated with the resource, and selectively recognizing the user seeking access to the resource based on a result of said comparing.
[0005] These and other implementations can include one or more of the following features. In some cases, the high information portion includes eyes and a mouth. In some other cases, the high information portion further includes a tip of a nose. Processing the captured image can include detecting a face within the captured image by identifying the eyes in an upper one third of the captured image and the mouth in the lower third of the captured image. The reference model includes a reference image of a face, and processing the captured image further can include matching the eyes of the detected face with eyes of the face in the reference image to obtain a normalized image of the detected face. Additionally, processing the captured image can further include vertically scaling a distance between an eyes-line and the mouth of the detected face to equal a corresponding distance for the face in the reference image in order to obtain the normalized image of the detected face. In addition, processing the captured image can further include matching the mouth of the detected face to the mouth of the face in the reference image in order to obtain the normalized image of the detected face.
[0006] In some implementations, comparing the processed captured image can include obtaining a difference image of the detected face by subtracting the normalized image of the detected face from a normalized image of a target face associated with a target profile. Comparing can further include calculating scores of respective pixels of the difference image based on a weight defined according to proximity of the respective pixels to high information portions of the human faces. The weight decreases with a distance from the high information portions of the human faces. For example, the weight decreases continuously with the distance from the high information portions of the human faces. As another example, the weight decreases discretely with the distance from the high information portions of the human faces. As yet another example, the weight decreases from a maximum weight value at a mouth-level to a minimum value at an eyes-line.
[0007] In some implementations, selectively recognizing the user can include presenting to the user a predetermined indication according to a user's profile. The resource can represent an appliance, and the methods can further include capturing the image using an image capture device of the appliance. Selectively recognizing the user can include turning on a display of the appliance, if the display had been off prior to the comparison.
[0008] In some implementations, processing the captured image can include applying an orange-distance filter to the captured image, and segmenting a skin-tone orange portion of the orange-distance filtered image to represent a likely presence of a face in front of the image capture device. Processing the captured image can further include determining changes in area and in location of the skin-tone orange portion of the captured image relative to a previously captured image to represent likely movement of the face in front of the image capture device. Also, processing the captured image further can include detecting a face within the skin-tone orange portion of the orange-distance filtered image when the determined changes are less than predetermined respective variations.
[0009] According to another aspect, the described subject matter can also be implemented in an appliance including a data storage device configured to store profiles of users associated with the appliance. The appliance further includes an image capture device configured to acquire color frames. Further, the appliance includes one or more data processors configured to apply an orange-distance filter to a frame acquired by the image capture device. The one or more data processors are further configured to determine respective changes in area and location of a skin-tone orange portion of the acquired frame relative to a previously acquired frame, and to infer, based on the determined changes, a presence of a face substantially at rest when the frame was acquired. Further, the one or more data processors is configured to detect a face corresponding to the skin-tone orange portion of the acquired frame in response to the inference, the detection including finding eyes and a mouth within the skin-tone orange portion. Furthermore, the one or more data processors are configured to normalize the detected face based on locations of eyes and a mouth of a face in a reference image. In addition, the one or more data processors are configured to analyze weighted differences between normalized target faces and the normalized detected face. The analysis includes weighting portions of a face based on information content corresponding to the portions. The target faces are associated with respective users of the appliance. Additionally, the one or more data processors are configured to match the face detected in the acquired frame with one of the target faces based on a result of the analysis, and to acknowledge the match of the detected face in accordance with a profile stored on the data storage device and associated with the matched user of the appliance.
[0010] These and other implementations can include one or more of the following features. The data storage device is further configured to store rules for analyzing the weighted differences including weighting rules and scoring rules, and rules for matching the detected face against target faces.
[0011] Particular implementations of the subject matter described in this specification can be configured to realize one or more of the following potential advantages. The techniques and systems disclosed in this specification can reduce the impact of lighting and emphasize skin variance. By acquiring images with the appliance's own image capture device, the approximate location and orientation of face features can be pre-assumed and can avoid the overhead of other face recognition systems. The disclosed methods can ignore face biometrics, and rather use feature locations to normalize an image of a test face. Further, the face recognition techniques are based on a simple, weighted difference map, rather than traditional (and computationally expensive) correlation matching.
[0012] The details of one or more implementations of the subject matter described in this specification are set forth in the accompanying drawings and the description below. Other features, aspects, and advantages of the subject matter will become apparent from the description, the drawings, and the claims.
United States Patent Application 20110317872
Kind Code A1
Free; Robert Mikio December 29, 2011
Low Threshold Face Recognition
Abstract
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, are disclosed for reducing the impact of lighting conditions and biometric distortions, while providing a low-computation solution for reasonably effective (low threshold) face recognition. In one aspect, the methods include processing a captured image of a face of a user seeking to access a resource by conforming a subset of the captured face image to a reference model. The reference model corresponds to a high information portion of human faces. The methods further include comparing the processed captured image to at least one target profile corresponding to a user associated with the resource, and selectively recognizing the user seeking access to the resource based on a result of said comparing.
Inventors: Free; Robert Mikio; (San Jose, CA)
Assignee: APPLE INC.
Cupertino
CA
Serial No.: 826581
Series Code: 12
Filed: June 29, 2010
Current U.S. Class: 382/103; 382/118
Class at Publication: 382/103; 382/118
International Class: G06K 9/00 20060101 G06K009/00
Claims
1. A method performed by an image processor, the method comprising: processing a captured image of a face of a user seeking to access a resource by conforming a subset of the captured face image to a reference model, the reference model corresponding to a high information portion of human faces; comparing the processed captured image to at least one target profile corresponding to a user associated with the resource; and selectively recognizing the user seeking access to the resource based on a result of said comparing.
2. The method of claim 1, wherein the high information portion includes eyes and a mouth.
3. The method of claim 2, wherein the high information portion further includes a tip of a nose.
4. The method of claim 2, wherein said processing the captured image comprises detecting a face within the captured image by identifying the eyes in an upper one third of the captured image and the mouth in the lower third of the captured image.
5. The method of claim 4, wherein the reference model includes a reference image of a face, and wherein said processing the captured image further comprises matching the eyes of the detected face with eyes of the face in the reference image to obtain a normalized image of the detected face.
6. The method of claim 5, wherein said processing the captured image further comprises vertically scaling a distance between an eyes-line and the mouth of the detected face to equal a corresponding distance for the face in the reference image in order to obtain the normalized image of the detected face.
7. The method of claim 6, wherein said processing the captured image further comprises matching the mouth of the detected face to the mouth of the face in the reference image in order to obtain the normalized image of the detected face.
8. The method of claim 5, wherein said comparing the processed captured image comprises: obtaining a difference image of the detected face by subtracting the normalized image of the detected face from a normalized image of a target face associated with a target profile; and calculating scores of respective pixels of the difference image based on a weight defined according to proximity of the respective pixels to high information portions of the human faces.
9. The method of claim 8, where the weight decreases with a distance from the high information portions of the human faces.
10. The method of claim 9, wherein the weight decreases continuously with the distance from the high information portions of the human faces.
11. The method of claim 9, wherein the weight decreases discretely with the distance from the high information portions of the human faces.
12. The method of claim 9, wherein the weight decreases from a maximum weight value at a mouth-level to a minimum value at an eyes-line.
13. The method of claim 1, wherein said selectively recognizing the user comprises presenting to the user a predetermined indication according to a user's profile.
14. The method of claim 1, wherein the resource represents an appliance, the method further comprising capturing the image using an image capture device of the appliance.
15. The method of claim 14, wherein said selectively recognizing the user comprises turning on a display of the appliance, wherein the display had been off prior to said comparing.
16. The method of claim 14, wherein said processing the captured image comprises: applying an orange-distance filter to the captured image; and segmenting a skin-tone orange portion of the orange-distance filtered image to represent a likely presence of a face in front of the image capture device.
17. The method of claim 16, wherein said processing the captured image further comprises determining changes in area and in location of the skin-tone orange portion of the captured image relative to a previously captured image to represent likely movement of the face in front of the image capture device.
18. The method of claim 17, wherein said processing the captured image further comprises detecting a face within the skin-tone orange portion of the orange-distance filtered image when the determined changes are less than predetermined respective variations.
19. An appliance comprising: a data storage device configured to store profiles of users associated with the appliance; an image capture device configured to acquire color frames; one or more data processors configured to perform operations including: apply an orange-distance filter to a frame acquired by the image capture device; determine respective changes in area and location of a skin-tone orange portion of the acquired frame relative to a previously acquired frame; infer, based on the determined changes, a presence of a face substantially at rest when the frame was acquired; detect a face corresponding to the skin-tone orange portion of the acquired frame in response to the inference, the detection including finding eyes and a mouth within the skin-tone orange portion; normalize the detected face based on locations of eyes and a mouth of a face in a reference image; analyze weighted differences between normalized target faces and the normalized detected face, the analysis including weighting portions of a face based on information content corresponding to the portions, and the target faces being associated with respective users of the appliance; match the face detected in the acquired frame with one of the target faces based on a result of the analysis; and acknowledge the match of the detected face in accordance with a profile stored on the data storage device and associated with the matched user of the appliance.
20. The appliance of claim 19, wherein the data storage device is further configured to store: rules for analyzing the weighted differences including weighting rules and scoring rules; and rules for matching the detected face against target faces.
Description
BACKGROUND
[0001] This specification relates to low threshold face recognition, e.g., a face recognition system that can tolerate a certain level of false positives in making face recognition determinations.
[0002] Most face recognition systems fall into one of two categories. A first category system tends to be robust and can tackle various lighting conditions, orientations, scale and the like, and tends to be computationally expensive. A second category system is specialized for security-type applications and can work under controlled lighting conditions. Adopting the first category systems for face recognition on consumer operated portable appliances that are equipped with a camera would unnecessarily use an appliance's computing resources and drain its power. Moreover, as the consumer portable appliances tend to be used both indoor and outdoor, the second category systems for face recognition may be ineffective. Such ineffectiveness may be further exacerbated by the proximity of the user to the camera, i.e., small changes in distance to and tilt of the appliance's camera dramatically distort features, making traditional biometrics used in security-type face recognition ineffective.
SUMMARY
[0003] This specification describes technologies relating to reducing the impact of lighting conditions and biometric distortions, while providing a low-computation solution for reasonably effective (low threshold) face recognition that can be implemented on camera-equipped consumer portable appliances.
[0004] In general, one aspect of the subject matter described in this specification can be implemented in methods performed by an image processor that include the actions of processing a captured image of a face of a user seeking to access a resource by conforming a subset of the captured face image to a reference model. The reference model corresponds to a high information portion of human faces. The methods further include comparing the processed captured image to at least one target profile corresponding to a user associated with the resource, and selectively recognizing the user seeking access to the resource based on a result of said comparing.
[0005] These and other implementations can include one or more of the following features. In some cases, the high information portion includes eyes and a mouth. In some other cases, the high information portion further includes a tip of a nose. Processing the captured image can include detecting a face within the captured image by identifying the eyes in an upper one third of the captured image and the mouth in the lower third of the captured image. The reference model includes a reference image of a face, and processing the captured image further can include matching the eyes of the detected face with eyes of the face in the reference image to obtain a normalized image of the detected face. Additionally, processing the captured image can further include vertically scaling a distance between an eyes-line and the mouth of the detected face to equal a corresponding distance for the face in the reference image in order to obtain the normalized image of the detected face. In addition, processing the captured image can further include matching the mouth of the detected face to the mouth of the face in the reference image in order to obtain the normalized image of the detected face.
[0006] In some implementations, comparing the processed captured image can include obtaining a difference image of the detected face by subtracting the normalized image of the detected face from a normalized image of a target face associated with a target profile. Comparing can further include calculating scores of respective pixels of the difference image based on a weight defined according to proximity of the respective pixels to high information portions of the human faces. The weight decreases with a distance from the high information portions of the human faces. For example, the weight decreases continuously with the distance from the high information portions of the human faces. As another example, the weight decreases discretely with the distance from the high information portions of the human faces. As yet another example, the weight decreases from a maximum weight value at a mouth-level to a minimum value at an eyes-line.
[0007] In some implementations, selectively recognizing the user can include presenting to the user a predetermined indication according to a user's profile. The resource can represent an appliance, and the methods can further include capturing the image using an image capture device of the appliance. Selectively recognizing the user can include turning on a display of the appliance, if the display had been off prior to the comparison.
[0008] In some implementations, processing the captured image can include applying an orange-distance filter to the captured image, and segmenting a skin-tone orange portion of the orange-distance filtered image to represent a likely presence of a face in front of the image capture device. Processing the captured image can further include determining changes in area and in location of the skin-tone orange portion of the captured image relative to a previously captured image to represent likely movement of the face in front of the image capture device. Also, processing the captured image further can include detecting a face within the skin-tone orange portion of the orange-distance filtered image when the determined changes are less than predetermined respective variations.
[0009] According to another aspect, the described subject matter can also be implemented in an appliance including a data storage device configured to store profiles of users associated with the appliance. The appliance further includes an image capture device configured to acquire color frames. Further, the appliance includes one or more data processors configured to apply an orange-distance filter to a frame acquired by the image capture device. The one or more data processors are further configured to determine respective changes in area and location of a skin-tone orange portion of the acquired frame relative to a previously acquired frame, and to infer, based on the determined changes, a presence of a face substantially at rest when the frame was acquired. Further, the one or more data processors is configured to detect a face corresponding to the skin-tone orange portion of the acquired frame in response to the inference, the detection including finding eyes and a mouth within the skin-tone orange portion. Furthermore, the one or more data processors are configured to normalize the detected face based on locations of eyes and a mouth of a face in a reference image. In addition, the one or more data processors are configured to analyze weighted differences between normalized target faces and the normalized detected face. The analysis includes weighting portions of a face based on information content corresponding to the portions. The target faces are associated with respective users of the appliance. Additionally, the one or more data processors are configured to match the face detected in the acquired frame with one of the target faces based on a result of the analysis, and to acknowledge the match of the detected face in accordance with a profile stored on the data storage device and associated with the matched user of the appliance.
[0010] These and other implementations can include one or more of the following features. The data storage device is further configured to store rules for analyzing the weighted differences including weighting rules and scoring rules, and rules for matching the detected face against target faces.
[0011] Particular implementations of the subject matter described in this specification can be configured to realize one or more of the following potential advantages. The techniques and systems disclosed in this specification can reduce the impact of lighting and emphasize skin variance. By acquiring images with the appliance's own image capture device, the approximate location and orientation of face features can be pre-assumed and can avoid the overhead of other face recognition systems. The disclosed methods can ignore face biometrics, and rather use feature locations to normalize an image of a test face. Further, the face recognition techniques are based on a simple, weighted difference map, rather than traditional (and computationally expensive) correlation matching.
[0012] The details of one or more implementations of the subject matter described in this specification are set forth in the accompanying drawings and the description below. Other features, aspects, and advantages of the subject matter will become apparent from the description, the drawings, and the claims.
SUroot
Extreme Android User
Damnit I was going to list that in my post. I know they all do it (sueing one another) but I think Apple do do it to an extreme. Its kind of off topic though.
This thread touts itself as the ultimate face off, but it never can be. Both OS's are constantly evolving, people like myself may inject things from a broader sense than the OS itself, and it is almost partly subjective. Added to which, there will be lots of these threads all over for years to come.
johnlgalt
Antidisestablishmentarian
This quickly degraded to a "Mine's better than yours" approach. I've used and continue to use all three, and I find each has its merit in different situations - however, I find that I deal more with Windows users than any for various reasons (which, if I were to list, would start another round of name calling) and that is that.
AngelA's certainly entitled to his / her opinion, as is everyone else. But the problem in all threads like these manifests when regular non-inciting language is dropped for words like "light years behind" and such.
Sorry, that's not a discussion anymore, that's pedagogical vitriol thinly veneered as an opinion. And that is why these discussions invariably get closed.
AngelA's certainly entitled to his / her opinion, as is everyone else. But the problem in all threads like these manifests when regular non-inciting language is dropped for words like "light years behind" and such.
Sorry, that's not a discussion anymore, that's pedagogical vitriol thinly veneered as an opinion. And that is why these discussions invariably get closed.
johnlgalt
Antidisestablishmentarian
Believe it or not, MacOS is actually very fun to use - I've never had as easy of a job of joining to the domain as I did when I joined all 27 of the iMacs in the lab to the AD-server on campus - it was the fastest joining of multiple machines I'd ever done.
All the Mac purists were all up in arms about rEFIt when I discovered it, many of whom were saying "I can just use the Command button at boot to pick my OS - Yes, but Windows users won't know to do that, so rEFIt serves its purpose well enough.
Also, for hackintosh, MakeUseOf has a guide for how to build a fully functional hackintosh... and I have Tiger and Leopard in my cache of discs here with me
All the Mac purists were all up in arms about rEFIt when I discovered it, many of whom were saying "I can just use the Command button at boot to pick my OS - Yes, but Windows users won't know to do that, so rEFIt serves its purpose well enough.
Also, for hackintosh, MakeUseOf has a guide for how to build a fully functional hackintosh... and I have Tiger and Leopard in my cache of discs here with me
SamuraiBigEd
Under paid Sasquatch!
I would like to try a hackintosh at some point, right now I want to get my first Linux machine up and running.
johnlgalt
Antidisestablishmentarian
That's easy lol - Linux is easy-peasy these days because of their own HAL that polls the hardware and then figures out which modules to load.
You want a really fast *nix box? Compile your own kernel and apps specifically for your hardware - so there is no guess work - it turns on, loads the modules it needs without polling, and poof- you're at a login in well under 30 seconds flat from power on.
Of course, that is not for the faint of heart lol....
You want a really fast *nix box? Compile your own kernel and apps specifically for your hardware - so there is no guess work - it turns on, loads the modules it needs without polling, and poof- you're at a login in well under 30 seconds flat from power on.
Of course, that is not for the faint of heart lol....
SamuraiBigEd
Under paid Sasquatch!
Not quite up to that, I am more of a hardware guy. Maybe with some serious hand holding I could accomplish it.