Jump to content

Talk:Facial recognition system

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Move details about the systems for particular countries to separate articles

[edit]

Would it be reasonable to shorten the article by moving the details for particular countries to individual articles? Somebody has created an article for India, though it is currently shorter than the section in this article. https://en.wikipedia.org/wiki/Facial_recognition_system#India_2 I have added a link - shall we move the text there? Oitio (talk) 23:11, 13 October 2023 (UTC)[reply]

Wiki Education Foundation-supported course assignment

[edit]

This article is or was the subject of a Wiki Education Foundation-supported course assignment. Further details are available on the course page. Student editor(s): Lindenq.

Above undated message substituted from Template:Dashboard.wikiedu.org assignment by PrimeBOT (talk) 21:02, 17 January 2022 (UTC)[reply]

Wiki Education Foundation-supported course assignment

[edit]

This article was the subject of a Wiki Education Foundation-supported course assignment, between 10 January 2020 and 23 April 2020. Further details are available on the course page. Student editor(s): Alexissphe.

Above undated message substituted from Template:Dashboard.wikiedu.org assignment by PrimeBOT (talk) 21:02, 17 January 2022 (UTC)[reply]

Wiki Education Foundation-supported course assignment

[edit]

This article is or was the subject of a Wiki Education Foundation-supported course assignment. Further details are available on the course page. Student editor(s): Bobbruffey.

Above undated message substituted from Template:Dashboard.wikiedu.org assignment by PrimeBOT (talk) 21:06, 16 January 2022 (UTC)[reply]

identifying smiles

[edit]

What is it called, identifying smiles, vs. frowns, vs. excitement, vs. someone sticking their tongue out? Where you're not so interested in identifying the individual, but in identifying the individual's expression --LionKimbro 00:49, 13 September 2005 (UTC)[reply]

I would say that it would be a "facial expression recognition system". violet/riga (t) 11:42, 13 September 2005 (UTC)[reply]
Maybe this is what you are looking for: Affective computing. --Scot.hale 15:38, 17 November 2006 (UTC)[reply]

Reliability

[edit]

The reliability of a "Facial Recognition System", as with any other system intended to compare physical characteristics of specified objects depends on at least the following factor:

  1. The amount of money to be dedicated to the development;
  2. The amount of time to be dedicated to the development;
  3. The allowable cost of the final equipment -- single-unit or serial production;
  4. The quality of the reference sample to be compared;
  5. The amount of time available for the comparison to be produced;
  6. The other circumstances of the comparison -- lighting, stealth, etc.

When these and other similar factors which have not been enumerated here are specified and accepted by all parties to the development, a system can be built to offer an arbitrary degree of satisfaction to the user.
Basically, it is all money & time.
It is how the world works: Even the Pharoahs understood this and acted on it, indeed they did, not to mention the erectors of the giant heads on Easter Island. —The preceding unsigned comment was added by 162.42.84.59 (talkcontribs) 04:56, 10 October 2006 (UTC)[reply]

Pharaoh's slaves and Easter Island's large stone heads weren't paid for by money.
Not to nitpick or anything, but the Pharoahs had a large slave population, far more than they needed. Also, if you are refering to the pyramids, that was done mostly by religious fanatics from some reports, and they probably weren't paid much at all, if anything. Just something to do when not busy in the farming seasons.
And did they use money or care about time at Easter Island? They didn't plan things out at all, which is why they ran out of trees, and then had to stop their insane building of giant stone heads for no reason, and find a new religion. Dream Focus 01:06, 13 December 2006 (UTC)[reply]
Uhh... thanks for that. 70.162.15.97 02:27, 22 March 2007 (UTC)[reply]
May I suggest you have a look at Wikipedia:Talk_page#Formatting? Now let's get back on topic... --oKtosiTe talk 20:05, 5 April 2007 (UTC)[reply]


This section seems to deal with the deployment issues of the system rather than its reliability and accuracy in recognising faces. darkblackcorner (talk) 12:20, 4 April 2010 (UTC)[reply]

Civil Liberties & Facial Recognition Software cited soure missing

[edit]

The source of Civil Liberties & Facial Recognition Software has disappeared. --oKtosiTe talk 20:00, 5 April 2007 (UTC)[reply]

US Visit

[edit]

What about US-VISIT and the new biometric passports? From what I understand, these are a major new application of face recognition. —Preceding unsigned comment added by 129.6.61.30 (talkcontribs) 23:06, 15 August 2007 (UTC)[reply]

References

[edit]

References:
3: State Agency Uses Facial Recognition Software to Fight Fake ID’s"
7: Willing, Richard. "Airport anti-terror systems flub tests Face-recognition technology fails to flag 'suspects'"
8: Civil Liberties & Facial Recognition Software.
Are dead. Fjejsing 07:03, 17 September 2007 (UTC)[reply]

Fixed 7 & 8, couldn't find an archive version of 3, so i chucked two new ones in. CaNNoNFoDDa 21:20, 17 September 2007 (UTC)[reply]

Consumer computer biometric

[edit]

The new Lenovo L3000 Y410 laptop has this feature. I have no idea how well it works nor how many other Lenovo laptops have this feature, but it totally freaked me out when I saw it. :) 205.157.110.11 21:15, 18 October 2007 (UTC)[reply]

I heard that the Lenovo/IBM tech is the same they had ten years ago [1]. There are other options as well, including a popular freeware application called LemonScreen available at KeyLemon[2] (formerly BananaScreen). I think there should be a separate section for consumer use of this technology as associated links and perhaps a comparison of apps (or maybe there is and I can't find it). Jason P Crowell (talk) 17:35, 22 January 2008 (UTC)[reply]

Interfaced with 'crowd control weapons' and tracking/targeting technology

[edit]

The purpose of which is to induce mental health claims of being attacked by 'radio waves', and also to covertly induce health symptoms in the victim, for marketing purpose, like nausea medication or increase business at special heart hosptitals (targeting those with health insurance). --71.127.23.22 (talk) —Preceding undated comment was added at 02:50, 17 September 2008 (UTC).[reply]

Er

[edit]

What happened to all the citations? —Preceding unsigned comment added by 75.220.230.216 (talk) 11:53, 24 October 2008 (UTC)[reply]

Retailing

[edit]

Does 'Future Developments-Retailing' deserve own section? There are many possible applications and IMO paying with face doesn't look very secure(instead of implanted chip/card). —Preceding unsigned comment added by 213.194.247.92 (talk) 18:19, 21 November 2008 (UTC)[reply]

It is fair to say that a Weakness of facial recognition for security is the fact that it could be so easily fooled by holding up a photo of the person you are trying to imitate.

(i.e.) in front of any camera.

Obviously not for airport scanners.

For a home PC system, for example; By holding up the Photo of the Family, on the desk, to gain access to the computer, it could be useful.

Edited by SRT. Courtesy of:

77.99.21.181 (talk) 16:05, 20 July 2009 (UTC)[reply]

Merger proposal

[edit]
Merge - Face Detection with Facial Recognition and 3D Face Recognition

The three articles are overlapping:

  • "Face detection is a computer technology that determines the locations and sizes of human faces in arbitrary (digital) images."
  • "A facial recognition system is a computer application for automatically identifying or verifying a person from a digital image or a video frame from a video source."
  • "Three-dimensional face recognition (3D face recognition) is a modality of facial recognition methods in which the three-dimensional geometry of the human face is used."

I suggest to merge these articles:

--Solphusion (talk) 21:42, 10 August 2009 (UTC)[reply]

  • Merge - If the articles were longer, I would understand a possible need for keeping them separate, but each is short enough that I do not think that it would be a problem to join them together into one article, especially since their topics are so similar and cover much of the same information. Untitled2 (talk) 02:54, 27 October 2010 (UTC)[reply]
  • Merge - Face Detection and Face Recognition should NOT be merged. Face recognition differs notably from face detection, involving identification of a face or faces, whereas face detection merely determines whether any face is present in a scene but does not determine the identity of the person detected. Accordingly, face recognition involves distinct algorithms versus face detection, and the uses for the two are distinct as well. I do agree that three-dimensional face recognition should be merged as a subset of face recognition. Grabbensac (talk) 17:48, 30 November 2009 (UTC)[reply]
I was alarmed to see that facial detection and facial recognition were proposed for a merge. These are two very distinct fields in computer science. Imagine proposing a merge between object oriented programming and dynamic programming. There is some overlap, and I'm not sure that there is dynamic programming without objects, but the two have very different implications and contexts. Facial detection is exactly what it sounds like (and what the description very accurately indicates), a process of determining whether faces are even present at all. Facial recognition assumes that a face is present and works to compare the face to a database of known faces. Not only are the data structures involved totally dissimilar, but the aims are totally different. I agree with the parent that 3D facial recognition could be merged. If the 3D recognition article was more informative I would be in favour of its existence, but it does not seem to contribute anything which does not fit here. Forgive the lack of a proper signature, but I am not a frequent wikipedian. I am simply surveying the state of a field I studied as a graduate student (which by no means makes me an expert, but certainly allows me to appreciate that detection != recognition). —Preceding unsigned comment added by 72.93.185.154 (talk) 01:15, 22 June 2010 (UTC)[reply]
  • DO NOT Merge - I agree with the previous comment; detection and recognition are two very distinct pattern-recognition problems; as a practitioner in this field with multiple patents granted I advise to keep the topics separated; if I can make some time I will try to add a little to each article to explain better the distinctions ... I'm just a bit busy at the moment editing a book on a related topic ... ;-) —Preceding unsigned comment added by 89.124.88.99 (talk) 09:33, 7 March 2011 (UTC)[reply]
Although there is currently some overlap between the articles, they are not all the same thing. Face detection does not require facial recognition, and neither require the facial expression capture. In fact, facial expression capture does not require face detection or facial recognition, either. As the lede implies, it isn't even necessarily video-based. Dancter (talk) 14:40, 7 October 2009 (UTC)[reply]
Don't they all capture your image somehow, by camera or picture, recognize various points on it, and use that to do something? Get a video game to respond to you, or use it to move a 3D character around. Once you have the information scanned in, you can compare it to recognize someone, recognize their expression, or just record their movements to move something else around the screen. Dream Focus 20:13, 7 October 2009 (UTC)[reply]
No. No image is necessary whatsoever for facial expression capture. You are describing just one approach. Even the image-based methods are not restricted to CV techniques. Dancter (talk) 02:54, 8 October 2009 (UTC)[reply]
I'm closing and removing, as nothing has happened in 20 months. That means the issue is stale and does not appear to have gained consensus. - Wikidemon (talk) 12:59, 13 June 2011 (UTC)[reply]

Software

[edit]

Has anyone used fotobounce ? Is it good?

  • Applied Recognition
  • Disclosure of Grant and Contribution Awards Over $25,000 Recipient Name: APPLIED RECOGNITION INC.
  • phillip grenier Vp Marketing at Applied Recognition Inc, Toronto, Canada Area]
  • Chris Studholme, PhD I am currently working with Applied Recognition as director of development for their Fotobounce application
  • Review
  • [3]"We do have plans to integrate other web apps into the AirSet environment and we are offering an API that we already have some developers working with, so there will be 3rd party software available both for your AirSet cloud computers as well as software that works with it. For example, a month ago we put out a joint press release with a company called Fotobounce. They are using our APIs to integrate their facial recognition photo management software with AirSet so you will be able to create AirSet photo albums and upload them to your cloud computer from the Fotobounce application. So stay tuned... "

gioto (talk) 01:33, 13 September 2009 (UTC)[reply]

A tutorial & open-source code for a realtime face recognition system based on Eigenfaces is available at [4] (note: I'm the creator). I think this link should be added to the "Software" list of the page, or atleast to the "External Links" section, since it lets people try out a Face Recognition System themselves rather than just read about it. Shervinemami (talk) 14:21, 14 June 2011 (UTC)[reply]

It mentions OpenBR as the only open-source facial recognition software but OpenBR is pretty much useless. It is a project sponsored by some strange not-for-profit yet affiliated with military defense (?) organization and has little to no documentation or support for actual real-world use (for example compilation instructions are only for Ubuntu and only for 13.06 which is a little restrictive to say the least). OpenCV (as used in the tutorial by Shervinemami above) has community support from the open source world and a large amount of quality documentation, and builds across all platforms relatively easily. Rather than the tutorial I'd say opencv should be included in the list of open source software (and probably mention OpenBR is low-quality abandon-wannabe-opensourceware). There are other open source projects (or projects closely related such as face detection) but I have no opinion on those. Peterjtracey (talk) 21:59, 15 July 2013 (UTC)[reply]

Clarity...

[edit]

"Popular recognition algorithms include Principal Component Analysis which eigenface, Linear Discriminate Analysis, Elastic Bunch Graph Matching fisherface, the Hidden Markov model, and the neuronal motivated dynamic link matching."

This doesn't feel like a sentence —Preceding unsigned comment added by 69.172.100.161 (talk) 05:10, 7 November 2009 (UTC)[reply]

Instead how about "Popular recognition algorithms include Principal Component Analysis (using EigenFaces), Linear Discriminate Analysis, Elastic Bunch Graph Matching (using Fisherfaces), the Hidden Markov model, and dynamic link matching."? darkblackcorner (talk) 12:25, 4 April 2010 (UTC)[reply]

Privacy concerns paragraph

[edit]

While I'm here, the privacy concerns paragraph: "Many citizens are concerned that their privacy will be invaded. Some fear that it could lead to a “total surveillance society,” with the government and other authorities having the ability to know where you are, and what you are doing, at all times. This is not to be an underestimated concept as history has shown that states have typically abused such access before.[21]" Should probably read: "Concerns have been raised about the potential impact of facial recognition technologies on privacy. One fear is that it will enable a “total surveillance society,” wherein a government has the ability to know all individuals are and what they are doing. Some occurrences of this phenomenon have already been documented.[21]"— Preceding unsigned comment added by 72.93.185.154 (talk) 02:24, 22 June 2010 (UTC)[reply]

Clinton Fookes

[edit]

I was concerned because I could not find any reference on Wiki to Clinton Fookes and/or his software despite the fact that the software won the Eureke Prize (people's choice). Perhaps someone should correct this! See http://www.abc.net.au/news/stories/2010/08/17/2985835.htm?section=justin

Oaec (talk) 22:14, 17 August 2010 (UTC)[reply]

This article is about the technology, not about people or the prizes they have won. HairyWombat 17:43, 30 June 2011 (UTC)[reply]

Robotics attention needed

[edit]
  • Check for accuracy
  • Check to see if merge should be done (check conversations above and assess)
  • Update if necessary
  • Reassess after work done

Chaosdruid (talk) 09:45, 16 March 2011 (UTC)[reply]

Disappearance of Madeleine McCann

[edit]

I removed the following from the article section Additional uses:

As part of the investigation of the disappearance of Madeleine McCann the British police are calling on visitors to the Ocean Club Resort, Praia da Luz in Portugal or the surrounding areas in the two weeks leading up to the child's disappearance on Thursday 3 May 2007 to provide copies of any photographs of people taken during their stay, in an attempt to identify the abductor using a biometric facial recognition application.<ref name=photographs>"Help find Madeleine McCann". Child Exploitation and Online Protection Centre. 2007-05-21. Retrieved 2007-05-21.[dead link]</ref><ref>Brown, David (2007-05-23). "We will travel anywhere to find Madeleine, say parents". London: Times Online. Retrieved 2008-06-02.[dead link]</ref>

Both the references for it are {{Dead link}}s and, frankly, I don't believe this. The police requesting people to send in their holiday photographs makes sense, but using a facial recognition system to "to identify the abductor" is silly. The technology would not be able to do more than the police simply looking at the photographs. If somebody has a <ref> which is not dead then we can restore the paragraph. HairyWombat 05:40, 1 July 2011 (UTC)[reply]

Regarding the reliability of fingerprints

[edit]

Fingerprints are evidently being challenged in many legal contexts as client's attorneys discover that much of the science supporting the accuracy of fingerprints for unambiguous identification has never actually been scientifically tested, and at least some proportion of fingerprint experts have no verifiable expertise, nor have many experts actually been tested for accuracy.

Here's one potential clarification from Wikipedia:

Fingerprint#Criticism Criticism of fingerprint accuracy

There are many more sources available on the web, including beaucoup sources from the Defense viewpoint.

Lee-Anne (talk) 21:42, 18 June 2014 (UTC)[reply]

Terminology: Face Recognition vs. Facial Recognition

[edit]

In the academic community, the term "face recognition" is more widely used than "facial recognition". For example, Google Scholar reports >3M hits for "face recognition" but only half that many for "facial recognition". Furthermore, the goal isn't to recognize a "facial" but to recognize a "face". Therefore, I suggest that we make "Face Recognition" the primary page and have "facial recognition" redirect to it. Thoughts? Tdietterich (talk) 02:16, 26 February 2016 (UTC)[reply]

I have edited the body of the change to implement this, but I have not created a new page under the name "Face recognition". Tdietterich (talk) 22:02, 19 December 2016 (UTC)[reply]

I would second this change. Notably, the Electronic Frontier Foundation uses "face recognition" (https://www.eff.org/pages/face-recognition). — Preceding unsigned comment added by 2606:A000:1126:4392:FD2B:C1D2:1D5C:7E90 (talk) 03:40, 4 October 2020 (UTC)[reply]

[edit]

Hello fellow Wikipedians,

I have just modified 2 external links on Facial recognition system. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

  • If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
  • If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 02:15, 29 December 2016 (UTC)[reply]

Error

[edit]

The article states that the Department of State holds a database of 117 million adults. However, none of the cited sources proclaim that the this database is created or maintained by the State Department. — Preceding unsigned comment added by 73.39.93.61 (talk) 01:24, 21 April 2017 (UTC)[reply]

Feelings

[edit]

After people took a look at the workplace where this took place, many assumed that the corporate world could sell this type of software under the visage `facial recognition software'. — Preceding unsigned comment added by 2602:301:7751:160:381C:3457:AEE3:8C00 (talk) 21:22, 24 August 2017 (UTC)[reply]

[edit]

Hello fellow Wikipedians,

I have just modified one external link on Facial recognition system. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

  • If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
  • If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 10:49, 27 September 2017 (UTC)[reply]

Question

[edit]

I am needing to know how and where and price on digital ID Tamtam199522 (talk) 11:25, 10 December 2019 (UTC)[reply]

Expanding on Bans

[edit]

Hello, I am planning on expanding the section on bans for a school project as there is only one sentence. I want to briefly expand on both San Francisco and Somerville. If anyone has any suggestions, please let me know! Alexissphe (talk) 18:59, 26 March 2020 (UTC)[reply]

Facial recognition under authoritarian regimes?

[edit]

I've seen examples of Ukrainian soldiers using FR to identify Russian war victims, but for some reason the issue of Russia, China, etc... using FR massively is not covered as well. I will probably extend the article about what I (as a Russian citizen) know about FR in Russia.

For example, as of Jan, 2019 there were[1] ~170 000 cameras in Moscow, most of them installed on residential building entrance doors. Saint-Petersburg has some too but at lesser scale. And so on. WildyLion (talk) 10:30, 18 June 2022 (UTC)[reply]

References

Outdated information tag

[edit]

This is an article that needs constant updating due to the rapid movement of AI, but in particular the entire Advantages and Disadvantages section is from the 2000s. That's really ridiculously outdated. Poketama (talk) 11:26, 30 June 2022 (UTC)[reply]

I added a section on AI based facial recognition, but was later removed because "Blog spam". The links were pointing to the original project pages of RetinaFace and ArcFace, and a blog explaining the different steps in an AI based facial recognition pipeline. Nuwiz (talk) 13:52, 31 August 2023 (UTC)[reply]
Your blog (and/or a vendor blog) are not usable sources on Wikipedia. See WP:RS and WP:SPAM. MrOllie (talk) 14:00, 31 August 2023 (UTC)[reply]
Most AI research is published on arXiv, in project pages, or in blog articles all these can be seen as forms of "self-published sources". If the Wikipedia want's to block those, it's OK, but I don't believe its a sensible approach for the particular case of AI. Nuwiz (talk) 14:11, 31 August 2023 (UTC)[reply]
Feel free to try to get the policies changed. But until that happens we'll be following them. MrOllie (talk) 14:13, 31 August 2023 (UTC)[reply]
Apart from the sources issue, your proposed addition seemed to me to fail to explain what difference AI actually makes to the facial recognition techniques already described in the article. Sbishop (talk) 15:59, 31 August 2023 (UTC)[reply]
Some addition could be made comparing the accuracy between AI approaches (ResNet / ArcFace) and traditional approaches; but that's the main thing: using AI models instead of manually crafted features / algorithms and getting better accuracy thanks to them. Nuwiz (talk) 16:08, 31 August 2023 (UTC)[reply]