[HN Gopher] Eye Contact Correction: Redirecting the eyes to look...
___________________________________________________________________
Eye Contact Correction: Redirecting the eyes to look at the camera
Author : thunderbong
Score : 150 points
Date : 2024-10-16 01:37 UTC (21 hours ago)
(HTM) web link (www.sievedata.com)
(TXT) w3m dump (www.sievedata.com)
| vintagedave wrote:
| The results here in their sample video look _really good_: other
| tech I've seen in the past looked "wrong". But the sample input
| is not one I'd characterize as looking away from the screen. Eyes
| move around like the person is thinking. The result video only
| looks more focused. It's effective in carrying focus (it really
| does matter when someone looks directly at you), but it's making
| tiny changes.
|
| > Limitations
|
| > Works best with frontal face views and moderate head rotations.
|
| > Extreme head poses or gaze directions may produce less accurate
| results.
|
| There it is. To use this I'd like to see an example showing it
| stop adjusting when "extreme" aka normal head poses are used. If
| it can handle real behavior and improve eye tracking in the
| optimal case so it's seamless adjusting / not as someone moves
| around, that would be a good product.
| EdwardDiego wrote:
| My issue is the usual with laptop cameras, if I'm looking at
| you, my eyes are looking downwards, and it's very awkward
| speaking into the camera without seeing your face as I speak.
| AStonesThrow wrote:
| This is unfortunate, and perhaps more pernicious than obvious
| deep fakes, is a video filter that lies to the recipients.
|
| Several years ago during the pandemic, I enlisted a job coach to
| get me hired. One of her paramount concerns was my eye-contact
| with the camera. She said it's so important. Am I paying
| attention? Am I an honorable man who maintains eye contact when
| I'm in a conversation? If I look away, am I collecting my
| thoughts, or prevaricating?
|
| Many supervisors, managers, and teachers will judge their
| employees by whether they can pay attention during meetings, or
| if they're distracted, in their phone's screen, looking at
| keyboard, glancing off at children or spouse. Even more
| important, if you're meeting your _wife_ and _she_ can 't even
| maintain your attention, what kind of husband are you?
|
| If you employ a gadget to _lie about this_ , then I hope they
| fire you and find someone who'll be honest. I hope your wife
| sends you to sleep on the sofa.
| karlgkk wrote:
| > If you employ a gadget to lie about this
|
| This has been enabled on iPhones, by default, for like 5 years
| now. You never even noticed.
|
| Their implementation only does a small adjustment, which works
| so well that most people don't even know it's being done.
| olyjohn wrote:
| If we never noticed it, do we even need it? I don't use
| FaceTime, but have never been bothered by where people are
| looking in any other video conferencing software.
| bravetraveler wrote:
| > You never even noticed
|
| I have seen three cameras in use in nearly a decade. They
| were all in interviews. I'm not avoiding opportunities,
| either. Legitimately 4+ hours a day
|
| Might be fair to say not many cared to see/be seen
| maximilianroos wrote:
| Sounds like the coach helped you maintain eye-contact with the
| camera. But if we get a tool to do this, then we're lying.
| Would you say the coach helped you lie?
| CGamesPlay wrote:
| That doesn't even make sense. The lie is that you're not
| doing the thing you are projecting as doing. You just said
| the coach helped the poster do the thing they projected as
| doing.
| function_seven wrote:
| I would go so far as to say the _un_ corrected gaze is a lie.
| When I'm on a videoconference, I _am_ looking directly at
| whoever is speaking, but the camera's physical placement tells
| the "lie" that I'm looking down at something else. This is
| because we haven't figured out a good way of placing the camera
| literally wherever the eyes of the other party show up on the
| screen. So the camera is, by necessity, in the wrong position
| for video conferencing. But if we can fix it in software, then
| we can mitigate the "lie" somewhat.
|
| This is especially true for my set up, where I have two screens
| side-by-side with the camera replaced right between them. I
| just stare at the camera because otherwise it looks like I'm
| looking way off to the left or right. If I do look at the
| people who are talking, what they see is me looking off at
| "something else." That's a lie! :)
| AStonesThrow wrote:
| This is true, and unfortunate, but for the past 100 years,
| everyone has known that to make eye contact with a camera,
| you look into its lens. The instantaneous display of output
| is very recent, and if you ask a professional actress or news
| anchor what they do in the studio, they will tell you that
| they're trained to look into the camera lens, no matter
| what's on the monitors.
|
| I contend that it's unproductive to train consumers
| otherwise. Yeah, we could look at the screen and have
| software correct it. Or, we may eventually integrate lenses
| into screens so that they're placed exactly right. But it
| seems kludgy to do this software fix. Just train people to
| look in the right place. (I hate iPhones and I'm
| unable/unwilling to do Facetime with them. Please use Meet or
| Teams.)
|
| I'm gradually building skills that let me be aware of what's
| on the screen without having to stare into it. Having a
| relaxed, wide field of vision helps with many things. Glasses
| are counterproductive here.
| Izkata wrote:
| News anchor yes, actor/actress no. They look off to the
| side.
| AStonesThrow wrote:
| Ah yes, that's true - unless the character is "breaking
| the fourth wall" like Clarissa Darling, they'll be
| avoiding the direct gaze of the viewer for sure.
|
| Another example, though, would be vocalists in a video;
| usually they'll be singing right at the viewer and making
| a connection there, unless they're just too cool and
| aloof.
| niij wrote:
| edit: studio_seven said it better than I could. You're confused
| on what the perspective is with videoconferencing. There is no
| hardware with a camera in the middle of the screen; so you're
| always "looking away" to some degree.
| AStonesThrow wrote:
| No, I'm not confused at all. As I pointed out, the standard
| for 100 years: if you want eye contact, you look into the
| camera lens. The only thing that's changed recently is the
| availability of a direct, instantaneous monitor to distract
| us.
|
| Furthermore, if this corrects _only_ someone who 's looking
| directly at the screen, it'd be tolerable. But does it also
| correct eyes looking at a keyboard, eyes looking at a
| smartphone screen, eyes looking at a wayward toddler? That's
| worse.
|
| Also... _ten cents per minute_? That 's highway robbery!
| niij wrote:
| If I'm looking at a camera lens I'm not making eye contact.
| This isn't about broadcasting it's about videoconferences.
| AStonesThrow wrote:
| No, you don't understand the definition of "eye contact".
| Contact, by definition, is when my eyes meet yours
| directly. It takes two to tango, and to maintain eye
| contact, it is necessary for both of us to cooperate.
|
| The camera is the eye. Anyone seeing video of me is
| seeing me through the eyes of a camera. Therefore, to
| "make eye contact" I look into the camera, not into
| arbitrary pixels. In videoconferencing, it's wholly
| irrelevant where my audience's eyes are located, whether
| they're even visible. In videoconferencing, our cameras
| are the eyes, and that's how to make eye contact, because
| when I see you on the screen looking into the camera,
| your eyes are directed towards mine seeing the screen.
|
| _For over a hundred years,_ any subject of a camera has
| known that if you look into that camera lens, then your
| gaze will be perceived as "eye contact" to any viewers.
| Where do you look when you're taking a selfie? Or a
| wedding photographer is taking your photo? Do you look in
| the photographer's eyes? Do you stare at his flashbulb?
| That's fucking nuts!
|
| Why is this so hard to understand?
|
| If AI is directed to help us lie about a particular, very
| human, interaction cue, then is it any surprise we're a
| world full of autists and Asperger babies?
| allenu wrote:
| That reminds me of a few months into the pandemic, one of the
| VPs at the company I was working at was presenting in a Zoom-
| based all-hands. I remember that he was very clearly looking
| directly into the eye of the camera as opposed to looking at
| his monitor's video feed like everyone else. I remember
| thinking that it felt a little bit weird and unnatural and very
| performative, like a politician, since he very obviously
| intentionally wanted to come across as more human by looking
| directly at the audience, although at the same time it was a
| fake look since he wasn't looking directly into the eyes of any
| one person, but a camera.
|
| Perhaps other people didn't think about it as deeply as I did
| and maybe it did have the intended effect, but I remember I
| didn't see him or anyone else doing the same thing in any
| future all-hands.
| Der_Einzige wrote:
| The fact that your feathers are rustled is what make it all the
| more delicious and delightful that it exists.
|
| All attempts by folks to subvert the freedom to direct one's
| attention where they want to are tyrannical in nature. If you
| can't detect it's happening, it effectively did not have a
| negative externality. The tree did not make a sound if no one
| heard it.
|
| This is the same thought that is used to justify not letting
| cashiers sit while they bag groceries. Those who think this
| love the taste of boots in their mouth.
|
| I hope that they fire those who refuse to get with the times on
| AI and embrace ludditism, and I hope your wife considers her
| future with you after the economic ruin that such practices
| will bring upon your family.
| Retr0id wrote:
| Does what it says on the tin, but honestly I find the
| "uncorrected" video more comfortable to watch.
| karlgkk wrote:
| There are other implementations that do a better job, such as
| Apple and Google's. They also are less willing to correct eye
| contact when it's "out of range" so to speak.
| hanniabu wrote:
| I think it's the lack of subtle movement, it's too strict and
| really locks the pupils front and center
| lloeki wrote:
| You mean frequent saccades?
|
| https://en.wikipedia.org/wiki/File:This_shows_a_recording_of.
| ..
|
| or the occasional look away? (for which there appears to be a
| feature for that)
|
| > Look Away: enable_look_away helps create a more natural
| look by allowing the eyes to look away randomly from the
| camera when speaking
|
| I expect both to be different: while saccades do happen when
| occasionally looking _away_ from a person, they also happen
| when looking _directly at_ one person because we don't
| constantly stare at a very specific unique and precise point
| on their face.
|
| For the demo video, try enable_look_away = true,
| look_away_offset_max = 10, look_away_interval_min = 1 and
| look_away_interval_range = 1 (then submit), which from the
| result I got should really be the default for a more natural
| result.
| ddfs123 wrote:
| I think it's just that naturally nobody is keeping 100% eye
| contact ( except maybe like TV news reporter ), it feels like
| an interrogation.
| XorNot wrote:
| This is what I realized is uncomfortable about camera on
| group meetings in teams - I can't mute _other people 's_
| video, and so it feels intensely weird to have a wall of
| people staring blankly at you.
| RheingoldRiver wrote:
| > I can't mute other people's video
|
| you can switch to another tab, use a miniplayer, in some
| apps u can focus one person's screen and if you choose
| someone who has a static avatar up you'll barely see other
| people's faces.
|
| The nuclear option is to install PowerToys [0] and put
| something always on top (im a fan of the hotkey
| winkey+space to toggle always-on-top on and off) in the
| exact position of the other video feeds. notepad or
| something.
|
| [0] https://learn.microsoft.com/en-us/windows/powertoys/
| abdullahkhalids wrote:
| This has been a feature since 2020 [1]. Similarly exists in
| zoom now.
|
| [1] https://answers.microsoft.com/en-
| us/msteams/forum/all/featur...
| prmoustache wrote:
| You can totally turn off incoming video on msteams. What
| you can't is have it as a default setting afaik.
| mvoodarla wrote:
| Original dev here. I tend to agree for this particular demo
| video as I'm reading a book and I don't blink in the original.
|
| The model tries to copy the blinks of the original video so
| it's possible that in other conditions, you'd notice less of
| this.
|
| Fun to see this feedback though, definitely something worth
| improving :)
| mlhpdx wrote:
| I likewise find the "corrections" uncanny. It's not just the
| one with the book.
| patrickhogan1 wrote:
| BTW your main site is throwing an error. Probably want to
| edit since your post is growing.
|
| https://www.sievedata.com/
|
| Application error: a client-side exception has occurred (see
| the browser console for more information).
| mvoodarla wrote:
| Original dev here. Unable to replicate this on my end, try
| refreshing?
| patrickhogan1 wrote:
| Interesting. The issue occurs because I have WebGL
| disabled, causing the createShader function you're using
| to throw an error. You can reproduce this by going to
| chrome://settings, disabling "Use hardware acceleration
| when available," refreshing the page, and then triggering
| the same error.
| isuckatcoding wrote:
| Cool but why...?
| deepfriedchokes wrote:
| Videoconferencing.
| karlgkk wrote:
| Apple does this on the iPhone, by default. When you're looking
| at someone's face on FaceTime, it modifies the position of your
| eyes to be looking directly at the camera - so the person on
| the other end sees you looking at them.
| s4i wrote:
| Really? We FaceTime a lot with my wife and I keep telling her
| that I can see her looking at her own face in the corner
| instead or me. Is that tech accurate enough to tell that the
| person is looking at themselves and not the other
| participant, and then not correcting the eyes if that's the
| case?
|
| Anyway, I'd much prefer if Apple didn't silently alter the
| eye direction of people calling me.
| karlgkk wrote:
| I think their adjustment is very minor and only happens
| when you're looking directly at the camera. It's minor
| enough that you've almost certainly seen it and not
| noticed. She may have the setting disabled, or she may be
| looking far enough away from the camera that it isn't
| triggering.
| richdougherty wrote:
| Kudos to the dev for coming up with the eye position fixing
| solution.
|
| Building further on this idea, I wonder if instead of changing
| the image to look at the camera, we could change the "camera" to
| be where we're looking.
|
| In other words we could simulate a virtual camera somewhere in
| the screen, perhaps over the eyes of the person talking.
|
| We could simulate a virtual camera by using the image of the real
| camera (or cameras), constructing a 3D image of ourselves and re-
| rendering it from the virtual camera location.
|
| I think this would be really cool. It would be like there was a
| camera in the centre of our screen. We could stop worrying about
| looking at the camera and look at the person talking.
|
| Of course this is all very tricky, but does feel possible right
| now. I think the Apple Vision Pro might do something similar
| already?
| mvoodarla wrote:
| This is an interesting idea. We are a little farther off from
| being able to do this but agree it would look really cool.
| scotty79 wrote:
| I think you'd get a lot by just transforming eyes so the gaze
| is relative to the virtual camera located on the screen at the
| place of the face of a person you are talking to. This way you
| get eye contact only when you are looking on their face on the
| screen, but not when you look somewhere else.
| newaccount74 wrote:
| There is already a lot of research on the 3D reconstruction and
| camera movement part, for example this SIGGRAPH 2023 paper:
| https://research.nvidia.com/labs/nxp/lp3d/
|
| In order for this to work for gaze correction, you'd probably
| need to take into consideration the location of the camera
| relative to the location of the eyes of the person on the
| screen, and then correct for how the other person is holding
| the phone, and it would probably only work for one-on-one
| calls. Probably need to know the geometry of the phone (camera
| parameters, screen size, position of camera relative to phone)
|
| Would be amazing, not sure how realistic it is.
| xnx wrote:
| Nvidia has free Broadcast software with an eye contact feature:
| https://www.nvidia.com/en-us/geforce/news/jan-2023-nvidia-br...
|
| It's from January 2023, so I don't know if they've improved it
| further since then.
|
| The video conferencing software providers have been way to slow
| to put whoever is speaking top-center (near where the camera
| typically is).
| zamadatix wrote:
| They also release it as an SDK these days so you can use it
| from tools like OBS https://www.nvidia.com/en-
| us/geforce/broadcasting/broadcast-...
| mvoodarla wrote:
| Original dev here. That's right, NVIDIA has a version
| available which we reference in our blog.
|
| https://www.sievedata.com/blog/eye-contact-correction-
| gaze-c...
|
| Newer models have come out that allow the same thing to be
| done and control even more than the eyes.
|
| See here: https://github.com/KwaiVGI/LivePortrait/blob/main/a
| ssets/doc...
|
| For web-conferencing, local use is great so NVIDIA's tools
| are what we recommend in that case.
| eisenman wrote:
| I appreciated using the Nvidia Tools for remapping webcam
| eye-contact until I was reviewing a recording and noticed
| that it changed my eye color. But it's been a bit. Perhaps
| an undocumented feature that newer versions/models fixed.
| Der_Einzige wrote:
| Thank you very much!
|
| - Signed, everyone whose currently trying to cheat on
| interviews that think that forcing peoples videos on does
| anything at all to keep them honest.
| EasyMark wrote:
| That is so creepy. I put the app up top center just under my
| camera (or whatever window I'm looking at and the focuses my
| eyes close enough to the camera to look like I'm talking at you
| (on the other side). I don't want some software interpolating
| my eye contact. Maybe I'm just ancient at 40 and easily creeped
| out by such things. The concept reminds of me of those
| paintings that seem to always be staring at you no matter the
| angle.
| boiler_up800 wrote:
| Looks really good and seems fast. My guess would be that this
| effect needs to be 99% or else people will notice something /
| although they may not be sure exactly what.
| thekevan wrote:
| From a development standpoint, this is cool.
|
| But the resultant video has a tad bit of uncanny valley going on.
|
| I'd rather learn from the guy on the right.
| mvoodarla wrote:
| Original dev here. Agree this video looks like uncanny valley
| but it's likely because the lighting of the original video is
| off + I baggy eyes (I was sleep deprived).
|
| Would recommend trying it on other videos, it is surprisingly
| good. Although there definitely are areas to improve.
| jpeggtulsa wrote:
| 10 cents per minute of video... Pass.
| albert_e wrote:
| Practically --
|
| I feel hardware technology can improve further to allow under-
| the-LED-display cameras .... so that we can actually look at both
| the camera and the screen at the same time.
|
| (There are fingerprint sensors under mobile screens now ...and I
| think even some front facing cameras are being built in without
| sacrificing a punch hole / pixels. There is scope to make this
| better and seamless so we can have multiple cameras if we want
| behind a typical laptop screen or desktop monitor.)
|
| This would make for a genuine look-at-the-camera video whether we
| are looking at other attendees in a meeting or reading off our
| slide notes (teleprompter style).
|
| There would be no need to fake it.
|
| More philosophically --
|
| I don't quite like the normalization of AI tampering with actual
| videos and photos casually -- on mobile phone cameras or
| elsewhere. Cameras are supposed to capture reality by default. I
| know there is already heavy noise reduction, color correction,
| auto exposure etc ... but no need to use that to justify more
| tampering with individual facial features and expressions.
|
| Videos are and will be used for recording humans as they are. The
| capturing of their genuine features and expressions should be
| valued more. Video should help people bond as people with as
| genuine body lanuage as possible. Videos will be used as memories
| of people bygone. Videos will be used as forensic or crime scene
| evidence.
|
| Let us protect the current state of video capture. All AI
| enhancements should be marketed separately under a different
| name, not silently added into existing cameras.
| yieldcrv wrote:
| Or buy a specialty device for replicating the real world
|
| Its been half a decade already from when I first noticed
| iphones cant capture a red world when wild fires are messing up
| the air quality, had to break out an ILC (DSLR without the SLR)
| to capture the world more congruently to how I see
| lloeki wrote:
| > iphones cant capture a red world when wild fires are
| messing up the air quality
|
| s/iPhones/the iPhone Camera.app/
|
| Apps like Halide and Pro Camera have no trouble handing you
| over control of white balance. I've captured both faint
| aurora borealis and red/brown hue when sand and dust is
| brought over to inland Europe by scirocco with great success.
| jrussino wrote:
| I agree with your philosophical stance, in general, but this
| particular use case is one that I've been wanting for years and
| where I think altering the image can be in some ways more
| "honest" than showing the raw camera feed.
|
| With an unfiltered camera, it looks like I'm making eye contact
| with you when I'm actually looking directly at my camera, and
| likewise it looks like I'm staring off to the side when I'm
| looking directly at your image in my screen.
|
| A camera centered behind my screen might be marginally better
| in that regard, but it still wouldn't look quite right.
|
| What I'd really like to see is a filter for video conferencing
| that is aware of the position of your image on my screen, and
| modifies the angle of my face and eyes to more closely match
| what you would actually see from that perspective (e.g. it
| would look like I'm making direct eye contact when I'm looking
| at/near the position of your eyes on my screen).
|
| You could imagine this working even for multiple users, where I
| might be paying attention to one participant or another, and
| each of their views of me would be updated so that the one I'm
| paying attention to can tell I'm looking directly at them, and
| the others know I'm not looking directly at them in that
| moment.
| wruza wrote:
| Would be funny if everyone on your screen gave a side eye to
| the bottom right corner where the currently speaking person
| is.
|
| Jokes aside, I think you're absolutely right. Online
| interactions have dynamic geometry, so mounting a camera
| behind a screen will just not cut it, unless the entire
| screen is a camera. Also, some people might prefer
| projecting/receiving no eye contact at all, at times, in
| situations. And vice versa.
|
| Philosophical stance here is purely traditionalist, it
| decides on behalf of people. What people would like to use,
| that should exist. "Videos are and will" is a strange claim,
| assuming its claimer has neither control over it nor any sort
| of affirmation that it is going to be true.
| albert_e wrote:
| Once we have technology to put a camera under a screen
| without sacrificing display quality ... we will not stop at
| one camera.
|
| There will be an array of cameras covering say every 2x2
| inch square of your screen.
|
| Just see how many cameras are on todays phones. Same can
| happen with new camera tech too.
|
| Also there will be a huge commercial driver to put multiple
| cameras under the screen -- all apps and marketers can
| track your precise gaze. Ads will pause unless you are
| actually watching them. I will hate it but it feels
| inevitable.
| iwontberude wrote:
| If that happened I would become a drastically different
| person because no one may control people's bodies like
| that except for themselves. God that really made me angry
| to read. I really really hope you are wrong.
| silver_silver wrote:
| Ads already pause if you switch apps on mobile, and
| vending machines/retail screens have had cameras in them
| for expression/attention tracking for years. It's not
| much of a leap from there
| hammock wrote:
| "Eye contact" is not a monolith though. Typically we look at
| someone's eyes when we are speaking but their mouth when they
| are speaking. And eye contact can be a pattern of crossing
| between their left and right eyes. And making and breaking
| eye contact are important parts of nonverbal communication.
| The typical AI "eye contact correction" will do none of this.
| redwall_hp wrote:
| It's also extremely culturally dependent. (Never mind that
| plenty of people in countries that obsess over eye contact
| find it uncomfortable as well.)
|
| It's generally considered rude or an act of intimidation to
| maintain eye contact with people in Japan, for example. Not
| nodding occasionally while someone is talking is also seen
| as a sign that you're not paying attention. Are we going to
| modify videos to nod automatically too? Or maybe we can
| stop trying to fake social interactions and enforcing local
| customs on the world.
| aitchnyu wrote:
| Will we have video with sensor signature for evidence purposes?
| One high court in India rejected any video evidence as a
| potential deepfake.
| prmoustache wrote:
| > One high court in India rejected any video evidence as a
| potential deepfake.
|
| Well I would have expected any court would have stopped
| accepting audio and video as evidence by now.
| agos wrote:
| that would be a lot of baby throwing along the bathwater
| ballenf wrote:
| Similar to how hearsay evidence is thrown out, despite it
| potentially having substantial value. Court is exactly
| the place you want to throw out the baby with the
| bathwater.
| JohnFen wrote:
| All I know is that I personally have stopped giving audio
| and video any benefit of the doubt. I think it's risky to
| accept any recording as representative of truth unless you
| or someone you trust was there at the time of the recording
| to vouch for its correctness.
| aspenmayer wrote:
| There are emerging standards for this.
|
| https://www.theverge.com/2024/10/15/24271083/youtube-c2pa-
| ca...
| sharpshadow wrote:
| So one could upload the original footage directly to
| YouTube to get the authenticity label and put it on private
| then proceed with the usual edits and provide the link to
| the private video as prove that it's real.
| aspenmayer wrote:
| Presumably that would break the chain of custody
| metadata, or would leave provenance breadcrumbs leading
| back to the original unedited video?
|
| https://support.google.com/youtube/answer/15446725?hl=en
|
| > Limitations
|
| > "Captured with a camera" only appears if a creator opts
| to use C2PA technology during filming. If it's missing,
| it doesn't mean the content has modified audio or
| visuals.
|
| > Note: This feature is separate from our existing
| altered and synthetic disclosures.
|
| > The metadata that leads to a "Captured with a camera"
| disclosure is made by a 3rd party (for example, a camera
| manufacturer). This means there is some risk that someone
| could take a photo of another screen showing synthetic
| content. Because the other screen shows an image that has
| been modified, it wouldn't be eligible for the "Captured
| with a camera" disclosure. This issue is called "air-
| gapping." Camera manufacturers will continue to develop
| detection measures to prevent "air-gapping," but the
| sophistication of those detection measures may vary in
| the near term.
|
| https://blog.google/technology/ai/google-gen-ai-content-
| tran...
| sadcherry wrote:
| > Videos are and will be used for recording humans as they are.
| The capturing of their genuine features and expressions should
| be valued more.
|
| Controversial stance, but for the same reason I reject wearing
| makeup.
|
| Girls, you are beautiful as you are! No need to fake it! Most
| guys don't do that either and everybody is perfectly fine with
| that too.
| irjustin wrote:
| The line is very long and blurry the whole way. The extremes
| are completely naked 100% of the time with zero grooming and
| the opposite is eugenics or genetic engineering body/facial
| features (is what i've come to believe?).
|
| Isn't it okay to feel good about looking good, sure (i love
| dressing up and doing my hair for occasions)! but obviously
| that can turn very problematic very fast. Honestly, I wish I
| knew where to draw the line in the sand. Is it makeup?
| piercings? nice clothes? surgery?
|
| Just a parent with two daughters who has more questions than
| answers.
| InDubioProRubio wrote:
| Surgery is permanent, life-long change- beauty, is relevant
| for 20years+
| ddingus wrote:
| What we did was draw the line at anything that might close
| a door in life they may prefer remain open.
|
| Messing with hair in our youth is fun and it grows back. No
| worries.
|
| Modest piercings society does not frown on. . No tattoos
| and especially none on the face, hands, etc...
|
| We had boys and girls and it went OK. Not too much
| complaining and when they became adults, we handed them the
| keys and wished them well and help where and how we can.
|
| Maybe our experiences help with understanding yours.
| HeatrayEnjoyer wrote:
| Please, please, tell me this is sarcasm.
| master-lincoln wrote:
| I don't think it was. And I agree: make up is like putting
| a mask on to hide who you really are because society taught
| you that you are more valuable this way. People might think
| they do this for themselves, but it has been put into their
| mind by media and adverts. This is not healthy and also
| wasted resources.
| botanical76 wrote:
| I partially agree with this, but at the same time... I
| don't feel that shunning the use of makeup and telling
| people their preferences are actually a result of
| societal brainwash is a good solution.
|
| If the problem is that society (in bubble X, Y or Z)
| teaches us our value is judged solely based on our
| appearance, then we should address the lessons we teach.
| I feel it is unproductive to play whac-a-mole with the
| emergent symptoms of such an underlying problem.
| ndndjdjdn wrote:
| Next up. Stop taking showers people!
| master-lincoln wrote:
| How is this a fair comparison? There are health benefits to
| hygiene, there are none from make-up
| wruza wrote:
| Yeah, looks don't get you anywhere in this world as a
| woman. /s
|
| ...We may talk all day how bad and unfair that is, but
| none of that changes the reality for an average person
| out there.
| Mountain_Skies wrote:
| What it really does is create an arms race within women,
| where those who opt out, with a few exceptions, are at a
| disadvantage to those who continue on and escalate. As a
| group, it would be a quality-of-life improvement for
| most, if they as a group ended the arms race but since
| there's no way for the group to enforce that, the arms
| race continues, with social media and technique videos
| advancing the front even further. For some the cost of
| participating in war becomes more expensive than the
| downsides, so they opt out and simply live with the
| disadvantage.
| hgomersall wrote:
| I'm not sure there are any established health benefits of
| showering routinely. Cleaning in response to
| contamination, sure, but every day with lots of soap etc
| I'm more sceptical of.
| genrilz wrote:
| This is actually probably more fair of a comparison then
| you'd think. Daily showers are bad for health your
| health[0], but I absolutely do them because my body
| produces a lot of oil and odor. This is a cosmetic reason
| similar to make-up.
|
| [0]: https://www.health.harvard.edu/blog/showering-daily-
| is-it-ne...
| master-lincoln wrote:
| You said daily, nobody prior did. Once a month is
| probably more beneficial than not at all...
| genrilz wrote:
| I feel like we might have been reading the original
| arguments in different ways, so let me summarize how I
| see this thread:
|
| I interpreted ndndjdjdn's comment as sarcasm. (due to the
| use of the phrase "next up") That is, I think he was
| saying that if you take sadcherry's logic to its limit,
| then people wouldn't shower or would shower less.
| sadcherry's logic is that people shouldn't wear make-up
| because it cosmetic is not beneficial to health. Thus I
| think ndndjdjdn was talking about the fact that people
| use showers for cosmetic reasons, and believes sadcherry
| probably doesn't actually want people to shower less, and
| so should probably rethink his views about make-up.
|
| You then posted your comment, saying that the health
| benefits of showers justify them even if they do have
| cosmetic benefits.
|
| I then comment, saying that I shower in a way that is bad
| for my health because of cosmetic reasons. I wanted to
| imply that a lot of people shower like this, and therefor
| the fact that moderate showers might have some health
| benefits is irrelevant, because the way many/most people
| shower is actually unhealthy. I probably should have been
| more explicit about the fact that I thought many/most
| other people shower in unhealthy ways.
|
| As an aside, I don't actually know of any concrete
| benefits to health besides making sure open wounds don't
| get infected. I tried to search the web for other
| benefits, and the only additional ones I got are
| exfoliation (which is cosmetic) and relaxation. (but
| relaxing things aren't generally classified as "healthy")
| With that in mind, I tend to believe the health benefits
| of showers are probably pretty over-hyped, (though not
| non-existent) and more like a cultural fiction to keep
| people showering than true knowledge.
|
| I'd be interested to hear if you have a different take.
| master-lincoln wrote:
| I agree on this, I previously didn't interpret the
| showering as a cosmetic action, but see that this line of
| thought would make sense now.
|
| To your aside of health benefits of showers: I also tried
| to research this, but other than getting rid of
| contamination (hazardous elements e.g. during
| construction or demolishing, or just dirt on wounds) I
| couldn't find any serious claim that washing the skin is
| beneficial for health (outside of making sure hands are
| clean before touching food or mucous membranes), I just
| assumed there should be one.
|
| I take my confident stance on this back...
| sadcherry wrote:
| There are parts of my body which, if not cleaned daily,
| will stink uncomfortably. Harvard webpage or not.
|
| I'd equate perfuming it over to make-up, not showering..
|
| It's also quite sad that a statement "we should put less
| make-up on" is immediately drifting into a discussion
| about not showering. Way to ridicule a viewpoint.
| sadcherry wrote:
| Guys who equate stopping to spend 30+ minutes a day
| painting your face with stopping to shower are part of the
| problem.
|
| It's exactly those unnatural expectations of looks that are
| put on women, starting at a really young age, that are the
| issue here. Not boys, just girls. It skews expectations and
| boom, everybody feels like they have to do it. It's very
| sad. I'm not saying don't shower, don't cut or even brush
| your hair, etc. All fine. But the full-on makeup you see
| walking through a random city in the morning, geez, what
| are we doing to ourselves. And what are the guys doing?
| Nothing close to it, but spend a lot of time justifying it.
| exitb wrote:
| Makeup is a personal preference. What OP talked about is
| subtly and transparently putting AI in a pipeline where we
| don't expect it. And it's not hypothetical, rather it already
| happens. Video meeting software is doing all kinds of sound
| rejection based on an unknown set of rules, even though none
| of us enabled that as a feature.
| tomooot wrote:
| It took me a couple of years to notice the "beautify"
| filter on my samsung S7 as I only ever activated the screen
| side camera by accident. When I did eventually use it a
| bit, I subconsciously knew something was off but assumed it
| was just spec differences between the two sensors and
| lenses, but then I noticed the "eyeball star twinkle" and
| realised what was up.
|
| On closer inspection it turns out it was actually smoothing
| my hair and boosting the contrast so I looked like I had
| dyed "highlights", along with airbrushing my cheeks a flat
| orangey coloured skin tone with a rosy center, as if I were
| wearing foundation and blusher!
| beeflet wrote:
| > Video meeting software is doing all kinds of sound
| rejection based on an unknown set of rules, even though
| none of us enabled that as a feature.
|
| It's optional on discord. Besides, it's conceivable that
| you might create a similar effect with a nice audio
| hardware setup
| voidUpdate wrote:
| Guys, you don't need to modify cars ever! They're fine as
| they are!
| DidYaWipe wrote:
| Are you seriously advancing that as a valid comparison?
| voidUpdate wrote:
| Yes
| sadcherry wrote:
| If guys do that to adhere to societal norms, then thats
| equally sad.
| JohnFen wrote:
| I fundamentally agree with you, but I wanted to mention the
| most common reaction I've got from women when discussing this
| topic: women mostly don't put on makeup for the benefit of
| men. It has more to do with societal expectation, setting
| social status with other women, and very often that they like
| that it's a mask.
| botanical76 wrote:
| I think the only case where a woman's use of makeup can be
| considered fake is when she lies about using it.
|
| Otherwise, it is just another way humans choose to dress
| their external appearance for their own pleasure, fulfilment
| and social intentions. It's not as if it's hard to tell when
| someone is wearing makeup - that is, at least when you're
| close enough to be able to inspect their imperfections at
| all.
|
| It seems to me that this idea about makeup being 'fake' stems
| from heteronormative dating, where a man may feel he is
| unable to properly assess a woman's beauty (and her
| attractiveness to him) if her face has been changed in
| arbitrary ways. But personally, I don't think we should
| optimize all human encounters for dating efficiency. More
| broadly, there is no social contract which stipulates you
| must wield your natural appearance at all times. I think we
| need not add more social expectations to an already long
| list.
| sadcherry wrote:
| The pure fact that there is an asymmetry between men and
| women w.r.t. makeup renders your argument void.
| YeahThisIsMe wrote:
| I agree with this.
|
| I don't actually want the person I'm talking to to appear to be
| looking directly into my eyes because it's weird - it means
| they're looking at the camera and not at me on the screen,
| talking to them.
| smeej wrote:
| Somehow I've apparently made a different adjustment to this
| than most people. My therapist was commenting on it the other
| day, how I do look directly into the camera when I want her
| to see me as making "eye contact," rather than looking
| directly at where _I_ see her eyes.
|
| She's taking this as an autistic adaptation NT people are
| less likely to make, like my gestures are practiced and
| tailored for the sake of the other, not my own sake. I want
| to "look in her eyes" to make a point, because that's one of
| the ways you show people you're making an important point,
| not to see how she's responding to what I'm saying.
|
| I haven't done any of it on purpose. It's apparently just how
| I've adapted to the weird communication space of having a gap
| between actually looking at someone's eyes and being _seen_
| to be looking at someone 's eyes.
| Iku_Tri wrote:
| I don't want to be mean to your therapist, but really?
|
| Understanding camera eyelines counts as autistic now?
|
| You're fine doing that. Sorry, but that comment she made
| really sent me.
|
| Reminds me of how the film department forced the digital
| artists to take a Cinematography and lighting classed irl
| so their final project renders would improve.
| smeej wrote:
| It might be one thing if I had done it on purpose,
| because I was thinking about camera eyelines. But it
| wasn't deliberate. I subconsciously choose based on how
| another person will see me, because I don't really expect
| to get a whole lot of information from seeing _them._
| Something about this being a type of "masking" in
| autistic women, trying harder to get my social cues
| across to others, but not expecting myself to receive
| them.
|
| I think maybe I have "trauma masquerading as ASD,"
| because the symptoms are subjectively improving as I
| learn to down-regulate my nervous system, but then I
| don't much care what label gets put on _why_ I 'm weird.
| I'm much more interested in figuring out what to do with
| the different _ways_ I 'm weird. I'm old enough that I
| can't think of ways formal diagnosis would _help_ me, so
| I 'd rather assume each challenge is treatable until I
| find out that it isn't.
| mannykannot wrote:
| Indeed - intense eye contact can be unsettling, even without
| the additional information gleaned from knowing that the
| other party has chosen to look at the camera.
|
| Eye contact is a subtle and important dynamic in human
| interaction (to the point where it has been suggested that we
| have white sclera, while our closest ape cousins do not, as
| an adaptation in support of easily detecting eye contact.) In
| a meeting, that includes third parties seeing who is making
| eye contact with whom.
|
| The systems being discussed here are too simple to restore
| this natural dynamic, and it is not clear to me that always-
| on eye contact correction[1] is free of unintended and
| undesirable consequences - for example, in some
| circumstances, it might ramp up the tension in a discussion,
| or it might help someone who is dissembling.
|
| [1] Even with random look-aways, I suspect - in actual
| conversation, look-aways are often correlated with what's
| going on in the discussion.
| boneitis wrote:
| > because it's weird
|
| I don't get many opportunities to express my exasperation
| with the paradigm of the youtube content creator's thousand
| video cuts per spoken sentence, but hell, in the same way, I
| think it's just $#@%ing weird.
| TowerTall wrote:
| > under-the-LED-display cameras
|
| If people laugh with their mouths open, wouldn't a camera
| placed below the LED display capture the inside of their
| mouths, and the rest of the time just point straight up their
| noses?
| albert_e wrote:
| I meant the camera will be invisible and BEHIND the screen
| .... just not visible as a punch hole/notch.
|
| I think some mobile phones have already done this...where
| they are able to put a camera behind the pixels.
| vitorsr wrote:
| > I know there is already heavy noise reduction, color
| correction, auto exposure etc ... but no need to use that to
| justify more tampering with individual facial features and
| expressions.
|
| Critically, the enumerated computational processing units are
| global transformations, while tampering is inherently a local,
| "contentful" transformation.
| jMyles wrote:
| > the enumerated computational processing units are global
| transformations, while tampering is inherently a local,
| "contentful" transformation.
|
| This is a brilliant way to examine / explain the distinction.
| xattt wrote:
| I always thought that under-screen cameras would come as a bug-
| eye lens, with the sensors between pixels. The pitch of modern
| mini-LED displays seems to have enough space between pixels to
| fit them in.
| IanCal wrote:
| I get the general philosophical point but to take a fun
| counterargument - cameras don't record the moment. They record
| a very narrow snapshot in time. One you, or anyone around you,
| may not recognise.
|
| Have you ever looked at a group of friends and thought "ONE OF
| YOU IS BLINKING"? No. Yet it's quite common to have a photo
| where at least one person is mid-blink. The 30-year lifespan of
| that photo includes the milliseconds they were blinking. Is it
| _untrue_ to have a picture where two people were not blinking
| and standing side by side? They did in real life, in those same
| poses, but fractions of a second apart. Is it a failure to
| capture reality by having a picture of them with their eyes
| open? Maybe - or maybe the blending of several moments is more
| true to the original situation than any specific snapshot could
| be.
|
| > I feel hardware technology can improve further to allow
| under-the-LED-display cameras .... so that we can actually look
| at both the camera and the screen at the same time.
|
| That doesn't fully solve the problem because you'd be looking
| at the middle of the screen not at the person talking to you in
| a group.
|
| > Video should help people bond as people with as genuine body
| lanuage as possible.
|
| I agree, but having people be able to actually look at each
| other is surely part of this.
| albert_e wrote:
| > That doesn't fully solve the problem because you'd be
| looking at the middle of the screen not at the person talking
| to you in a group.
|
| Repeating my comment on a sibling ...
|
| Once we have technology to put a camera under a screen
| without sacrificing display quality ... we will not stop at
| one camera. There will be an array of cameras covering say
| every 2x2 inch square of your screen.
|
| Just see how many cameras are on todays phones. Same can
| happen with new camera tech too.
|
| Also there will be a huge commercial driver to put multiple
| cameras under the screen -- all apps and marketers can track
| your precise gaze. Ads will pause unless you are actually
| watching them. I will hate it but it feels inevitable
| vlovich123 wrote:
| Honestly I'll take the software correction approach. Seems
| cheaper. I'll also challenge about whether people actually
| care about the philosophical position about live editing.
| Zoom filters to ade makeup and other realtime and non
| realtime filters are popular. Movies have special effects.
| I think this purism isn't helpful given what it seems that
| people actually want, not to mention that the concept of
| "true image" is so tenuous (eg no picture of the aurora
| borealis or the Milky Way is actually what your eye would
| see).
| ballenf wrote:
| Artwork can be more true to reality than a photograph.
| JohnFen wrote:
| > cameras don't record the moment. They record a very narrow
| snapshot in time.
|
| Isn't a "moment" a very narrow snapshot in time by
| definition?
| wizzwizz4 wrote:
| Colloquially, "the moment" also includes the context, both
| immediate and general.
| mikae1 wrote:
| _> I feel hardware technology can improve further to allow
| under-the-LED-display cameras_
|
| Everything but your smartphone is big enough that you'd to
| sprinkle your entire screen area with sensors to get the sense
| of me looking at you. And, _that_ won 't be cheap.
|
| Say my laptop had a sensor dead center and I was in a group
| chat. Only the person dead center would see me looking to the
| camera.
|
| This is better done in software.
| JohnFen wrote:
| > I don't quite like the normalization of AI tampering with
| actual videos and photos casually -- on mobile phone cameras or
| elsewhere.
|
| I agree. This is one of the things that I actively worry about.
| taeric wrote:
| Beam splitting is a thing. Elgato has a lowish cost one that
| works quite well.
| renewiltord wrote:
| The really sad thing is that we take raw sensor data and
| process it at all. People are so out of touch with things these
| days we use lenses to focus the picture etc. Why not just
| transmit the raw sensor data instead of processing everything
| so much? People could just use their minds (I know, ridiculous
| to ask people to do that in this era where everything is spoon
| fed to you) and actually interpret things for once.
|
| What a society! Processed food, plastics in their blood,
| processed sensor data. Ugh, we have strayed so far from natural
| interactions.
|
| Philosophically we have abandoned being mindful of where we
| are, and just being our natural forms instead of being slaves
| to what some computer is telling you that you should be seeing.
| jpicard wrote:
| Here's a webcam that has a small arm that drops down placing
| the sensor in front of the screen. It blocks a little bit of
| the screen, but allows more eye contact, without using AI to
| modify eyes.
|
| https://icontactcamera.com/
| not_a_bot_4sho wrote:
| I've never seen an implementation of this that wasn't super
| creepy past the initial tech demo
| lelandfe wrote:
| iOS Facetime quietly does this by default, I don't notice it
| Applejinx wrote:
| I put a monitor screen behind my camera for video-making,
| adjusted so the eyes are just barely showing over the camera.
| Then, when my eyes are drawn to me on the screen, I'm looking
| at the camera (just over it) which works pretty well. The eye
| contact is pretty good, but I can look away or around all I
| want: I'll just tend to be drawn to 'display of eyes' so I'm
| hacking that instinct to make better videos.
|
| That said, plenty of people don't make eye contact with the
| camera much at all :)
| throwaway290 wrote:
| I would pay for the opposite.
| advisedwang wrote:
| This is the real killer feature of Google's project starline,
| although they also achieve a 3D display.
| AyyEye wrote:
| Only a techbro would think that "eye contact" means just
| synthesizing eyes. It's a high bandwidth communication medium and
| synthesizing it removes what little we had. Yes I know this isn't
| the first, no I don't think any of this reality-meddling is any
| less creepy.
| patrickhogan1 wrote:
| Really cool application.
|
| Just a heads up - your main website is showing an error. You
| might want to fix it since your post is gaining traction. Here's
| the link: https://www.sievedata.com/
|
| The error message reads: 'Application error: a client-side
| exception has occurred (check the browser console for more
| details).'
| mvoodarla wrote:
| Original dev here. Unable to replicate this on my end, try
| refreshing?
| patrickhogan1 wrote:
| Interesting. The issue occurs because I have WebGL disabled,
| causing the createShader function you're using to throw an
| error. You can reproduce this by going to chrome://settings,
| disabling "Use hardware acceleration when available,"
| refreshing the page, and then triggering the same error.
|
| Sorry for duplicate post. Also this feature is enabled by
| default, but causes issues with several sites.
| ta8645 wrote:
| Great. Now, do the rest of me sitting in the seat. If you don't
| need my real eyes, you don't need any of the real me. We can
| discuss, whatever it is, in email.
| Bengalilol wrote:
| I, for some reason, prefer the original video. I may have an eye
| contact problem. Otherwise, the feature is nice and almost
| perfect: there could be some spaces where eye contact shouldn't
| be always on, I bet this would make it more human.
| DidYaWipe wrote:
| Creepy and misguided. Do people stare at you fixedly and
| unwaveringly during in-person conversations?
|
| And if they do, do you like it?
| lloeki wrote:
| > Look Away: enable_look_away helps create a more natural look
| by allowing the eyes to look away randomly from the camera when
| speaking
|
| For the demo video, try enable_look_away = true,
| look_away_offset_max = 10, look_away_interval_min = 1 and
| look_away_interval_range = 1 (then submit), which from the
| result I got should really be the default for a more natural
| result.
| qwertox wrote:
| Usually looking away is part of a gesture which involves the
| context, like facial muscles and the information being shared
| ("Hmm, when was this?": makes the eyes looks up)
| dTal wrote:
| Okay, these options are far enough down the slippery slope to
| present a compelling argument that the whole thing is a Bad
| Idea. Short hop from here to suppress_yawns=true, and then on
| to enthusiasm_multiplier=1.4...
| HKH2 wrote:
| If you mean someone listening is actually staring (without
| changing their facial expressions), then of course that's
| weird, but if they are adjusting their facial expressions to
| show reactions without looking away, what's wrong with that?
| JohnFen wrote:
| > if they are adjusting their facial expressions to show
| reactions without looking away, what's wrong with that?
|
| A whole lot. Even if they have varying facial expressions,
| not looking away is creepy as hell because looking away
| during conversations is actually an important aspect of the
| communication. Not looking away is sending a nonverbal
| message, and none of the usual ways that's interpreted are
| positive.
| blkhawk wrote:
| During corona I build a fold down thing that put my webcam at
| eye-level on my monitor. turns out with a large enough monitor it
| really isn't that bad to have the camera in front of it.
| drewbitt wrote:
| There was a Kickstarter-esque product at the time that did this
| too. Tiny camera that dangles down in front of the screen.
| jedisct1 wrote:
| Doesn't FaceTime already do that?
| kleiba wrote:
| It would be great to have a feature where my closed eyes are
| replaced with open eyes looking at the camera - then I could
| sleep through boring meetings.
| chucksmash wrote:
| > It would be great to have a feature where my closed eyes are
| replaced with open eyes looking at the camera
|
| There's already a feature that does this called HR
| ClassyJacket wrote:
| It's just a matter of time until half of us are having an LLM
| deepfake voice clone of us attend "meetings" so we don't have
| to.
| boomskats wrote:
| The thing with eye contact, though, is that it is worthless if
| you are never able to look away. When it's artificial like this,
| it's worse than not being there at all. It's just creepy. It was
| the same with nvidia's implementation a couple of years ago. It
| was just weird.
|
| I do appreciate that this is a problem worth solving though, and
| I spent a lot of my time during COVID worrying about the negative
| impact that normalising loss of eye contact would have on the
| social interactions of our younger generations.
|
| Back in 2021, I took one of those PS50 teleprompter mirrors that
| YouTubers use, put a 7in raspberry pi display in the slot where
| you're meant to put your phone, and made it my 'work calls
| display' for a couple of days. The interesting thing is that the
| only people that noticed without me pointing it out were
| completely non-technical, and when they did they complemented me
| on the quality of my webcam rather than the fact I was looking
| straight at them; they could tell something was better, but
| couldn't quite put their finger on it. Which is funny because I'm
| sure being stuck behind a cheap perspex one way mirror made my
| actual camera quality a bit worse.
|
| I remember I got to the point where I started playing with cv2
| trying to do realtime facial landmark detection on the incoming
| feed and having a helper process shift the incoming video window
| around the little screen so that it would keep the bridge of the
| other person's nose (the point I naturally made eye contact with)
| pinned to the bit of the screen that was directly in front of the
| webcam lens. Then one morning I walked into my office, saw this
| monstrosity on my desk, realised I was nerd sniping myself and
| gave up.
|
| One thing I do remember though is how odd it felt looking at
| yourself in a mirror without your image being mirrored. Not sure
| my brain was ready for that one after thousands of years of
| looking at itself in mirrored surfaces.
|
| Bit of a weird pic but the only one I can find:
| https://pasteboard.co/BXE6zhbpOD7E.jpg
| blitzar wrote:
| I wanted to do this but got stuck in the rabbit hole of picking
| out telepromters, screens and sizes. In the end my solution was
| to mount my webcam in the middle of the monitor (with the other
| party partially obscured). Previously my technique was to look
| at the camera not the screen (or have the other party in a very
| small window at the top of my screen) so partially obscured is
| an improvement!
| lloeki wrote:
| > One thing I do remember though is how odd it felt looking at
| yourself in a mirror without your image being mirrored. Not
| sure my brain was ready for that one after thousands of years
| of looking at itself in mirrored surfaces.
|
| Feynman has a good explanation for that:
| https://www.youtube.com/watch?v=msN87y-iEx0
|
| But it doesn't go deeper as to why we're perceiving ourselves
| that way, for that we have to dive into biology, neurology,
| bilateral symmetry, and the fundamentals as to how, as
| bilaterally symmetric beings, we're able to orient ourselves in
| a 3D world.
|
| (I recall reading a paper or watching some video about that,
| but can't find it anymore)
| lemonad wrote:
| I can definitely see the use case as it is annoying having to
| choose between actively looking at participants of a meeting on
| screen or _appear_ to look at the participants by gazing into the
| camera and not actually looking at them.
|
| I sometimes use an Elgato Prompter to better enable eye contact
| during meetings. The camera and lens is mounted behind the screen
| so looking at the screen is also looking at the participants. The
| downside is that the screen is tiny and you leaning forward to
| read, say, a document does not look that great on camera. So
| either you have to zoom it substantially or read it on another
| screen, thus looking away from the participants. In this case
| though, you are not looking at the participants and faking that
| eye contact in this case would be kind of weird.
| asdff wrote:
| Somehow the idea of everyone looking at the camera to wave
| goodbye, while in the process only seeing the camera and not
| the people you are trying to make virtual eye contact with, is
| hilarious to me. Like some dystopian comedy.
| jbverschoor wrote:
| Isn't this built in FaceTime?
| qwertox wrote:
| Looks somewhat creepy.
|
| The normal thing is not to uninterruptedly look at a person
| (which the camera is supposed to be). For example when you make a
| gesture of trying to remember something by looking somewhere
| else.
| 123pie123 wrote:
| it's definately in the uncanny valley for me
|
| he turns his head a little and his eyes look wrong
| EasyMark wrote:
| The creepy portrait that watches you no matter what part of the
| room you observe it from. For example the two in this jpg
|
| https://publish.purewow.net/wp-content/uploads/sites/2/2024/...
| nicholasbraker wrote:
| I assumed this was a roadmapped feature on (at least) Facetime on
| OSX/IOS. I never saw any implementation of it, but I see value in
| such feature. Also for Teams etc.
| theodorton wrote:
| It's available in Settings > FaceTime > Eye Contact.
| nicholasbraker wrote:
| Thanx!
| froh wrote:
| wow that's creepy :-)
|
| technically cool, however I'd rather prefer some semi transparent
| mirror set up.
|
| such a set up keeps the eyes alive.
| rzzzt wrote:
| DIY Perks has a build log on such an optics-based system:
| https://youtu.be/2AecAXinars
| kmfrk wrote:
| We've come a long way from red eye correction.
|
| I think it's great that this is labelled as "correction" as in a
| means of optional postprocessing when it's convenient. Nvidia
| implying that it's something we should enable by default rubs me
| the wrong way, but then again, I don't spend my day stuck in
| virtual meetings.
| krosaen wrote:
| This is really cool, glad to see someone making a real product
| out of this - have seen demos in the past.
|
| One thing I've always wondered is if this could be made to work
| for group video chats - depending on the tile you are looking at,
| that person would know, so you could tell who is paying attention
| to you, or even exchange a furtive glance with a colleague in
| reaction to someone else said like IRL. Even harder, but also
| cool would be updating the gaze dynamically so you could tell
| what they were looking at in your scene - say you have a
| whiteboard behind you and you can tell when the person is making
| eye contact with you vs looking at something you drew on the
| board.
|
| Original dev make it so! :)
| Arch-TK wrote:
| What is the modern obsession with video calls?
|
| I have been working for a company which allowed full remote work
| without any qualms since before COVID and nobody did video calls
| back then. Since we end up on site in secure environments we also
| just get told to disable the camera in the BIOS as part of our
| laptop hardening.
|
| For things like bi-annual meetings with your manager you would go
| into your local office.
| nonameiguess wrote:
| I don't remotely get it, either. Most group calls I'm in
| involve one person presenting something and we're watching
| that, not each other. For things like a one-on-one call with my
| manager, we just talk. Shit, sometimes we even talk on the
| phone, but usually still a computer. He might be wearing no
| pants for all I care. I usually have a cat on me and am
| probably laying flat. If it's before 10 AM, because my wife
| goes to work so late, she's probably walking around naked
| behind me and nobody needs to be seeing that.
|
| My company was also 100% remote from its start, even before
| Covid.
|
| As others have stated, it's also just unnerving to have people
| making nonstop eye contact or staring at any part of your body
| at all, even if it isn't your eyes. Maybe 90s Los Angeles was
| an abnormally shitty place to grow up because of the gang
| activity, but this is the kind of thing kids started fights
| over all the time. Robert DeNiro's most famous movie scene is
| about how threatening it feels to have someone looking at you.
|
| This isn't even unique to humans. When you regularly interact
| with animals, you're taught to look away and not hold direct
| eye contact because they'll see it as a challenge or threat.
| I've learned to do this with my own cats to make them more at
| ease. You learn to blink, narrow your eyes, look to the side.
| asdff wrote:
| Not only that but the obsession with higher and higher
| resolution video calls. Lets upgrade the laptop cams. No wait
| lets use a mirrorless camera and a ring light. Meanwhile the
| upload speed from your home isp is still early 2000s tier
| speeds and zoom downsamples you to 144p.
| schmichael wrote:
| Also a fully remote company worker since pre-COVID, and I
| vastly prefer calls with the video on. It's the norm at our
| company. Facial expressions convey enormous amounts of
| information. Also our team silently claps for people which is
| admittedly quite silly but hey, we like it.
|
| I'm rarely on calls without video, but when I am I find it
| jarring when voices just appear out of the ether with only a
| little flashing icon to indicate who it is I'm listening to.
|
| To each their own!
| fleischhauf wrote:
| having worked with half a remote team from Poland which also
| never turned the camera on I can tell you that faces provide
| a lot better method of person disambiguation for Germans than
| polish names. (no disrespect to the Polish language or
| people, my lack of knowledge therof is to blame here)
| EasyMark wrote:
| It's bizarre, I recently had an interview where the 2 people
| before and myself just did voice + coderpad over teams, no
| video. Then the next person, also with no video requested I
| open up my camera, I politely asked that they do the same if
| were to do it and they refused, so I refused. They said the
| interview was over. I was like fine, have a nice day to their
| very rude ending of the interview. It's strange how people
| expect to have all the power (or video?)in these situations.
| The hiring manager called me later and got my side, then
| apologized and asked if we could set up the 3rd interview with
| another team member, but I politely declined and went on about
| my day.
| neveroddoreven wrote:
| I wonder how soon we'll just be rendering real-time AI avatars of
| ourselves that are traced by our facial movements. Don't have to
| worry about fixing your hair or lighting or wearing nice clothes;
| just render whatever looks most appealing with a model.
| alexsmirnov wrote:
| Some decades ago, I worked with professional photo artists. Some
| did make photos for fashion magazines, one won "Word press photo"
| award couple times. When they make portraits, one rule is to
| always ask model to look slightly off the camera. As some
| explained, it adds some "life" to portrait. So, the tool does
| just the opposite to the art rules.
| EasyMark wrote:
| Pro-tip, put something interesting near wherever your camera is
| and look at that. I usually put the window itself (or whatever
| window is most interesting) up at the top center and that will
| get your eyes close enough to emulate "eye-contact"
___________________________________________________________________
(page generated 2024-10-16 23:01 UTC)