Recently apps like ‘Be My Eyes’ and ‘Seeing AI’ have added the ability to generate image descriptions using multi-modal Large Language Models (which can understand and generate both text and images) like GPT4Vision. I’m sure everyone has tried them, but for anyone who hasn’t, here’s the description I got by sharing a Toot to Be My Eyes this morning:
“The picture you've shared is a beautiful watercolour painting of flowers. The flowers have delicate, ruffled petals in shades of pink, purple, and white. The center flower is particularly striking with deep magenta at the base of its petals, which fades into a lighter pink. The leaves are painted in soft greens with hints of yellow and blue. The background is white, which makes the colors of the flowers and leaves pop. In the bottom right corner, there is a signature that reads "Wendy Craig" with the word "Photographer" underneath. The painting has a very soft and elegant feel to it.”
I have no way of knowing if the original image looks anything like this. But it doesn’t really matter, it fits in with what I was expecting from the Toot. I’ve really enjoyed ‘looking’ at all the Caturday photographs without Alt Text today – I can’t lie, I love cats!
Now, we all know the ‘scene description’ in Seeing AI was complete and utter rubbish, I don’t think that is me being harsh. The descriptions you get out of these new apps/models are at the least, very entertaining. At the best they are life changing!
For the first time, I, we, can ‘almost’ be a part of the visual internet. I’m not quite using Instagram daily, but you can Share post direct to Be My Eyes there to and it’s as good as you’ve seen.
I’m very excited, but I’m also aware that there are risks. These include inaccurate or misleading descriptions, so nobody should solely rely on this technology for critical tasks. But it is only going to get better. There is such an incentive for the world to develop computer vision and we will benefit. And it is only ever going to be better than it is right now, which is pretty fantastic.
So, here’s my question – are you as excited as I am? Are you keen to jump into the visual internet? Does the idea of GPT5Vision working with video make you want it to be here now? It’s certainly made using apps like Instagram, Ivory and Threads more inclusive.
Finally, on a personal note. I could see up to the age of sixteen. I’m using the text to image features of these models, together with the image to text, to create my own art. The descriptions you get from Google’s Bard are quite different than those from Be My Eyes and seeing AI – the differences can be quite interesting.
Comments
One day
I believe one day scientists will figure out the artificial retina thing they have been working on the past decade or so. I believe the webcam implants will lead to something greater.
I also believe one day eyes and optical nerves will be repairable/replaceable or otherwise worked around so the sightless can one day, or once again, see the world.
I realize the comment above does not necessarily have to do with AI, except to say that technology is growing. AI tech is growing. So imagine what AI and technology in general will be like in 20 years.
10 years?
Maybe tomorrow?
I know I am excited for the future of technology and AI. How about the rest of you? 😀
Finally, since we are sharing here, I had perfect or near perfect eyesight for the first 33 years of my life. I am in my late 40s now.
Touched a computer for the first time in the late 80s, anyone remember the old Apple II's?
Received my first PC in 1995, running Windows 95. Taught myself how to use it and eventually design webpages.
Used a MacBook and iPhone 4S for the first time in 2012, and taught myself how to use those as well. Of course I was fully blind at this point.
Eventually I went through training at a 'center for the blind' and then began my college career shortly after. Eventually graduated with a bachelors in Management Information Systems.
I am not saying this to sound important or anything, but if I could do all of the above post-blindness...
Imagine what I could do "with" eyesight. 🙃
Art
As a fairly unsuccessful artist, I look forward to using these AI picture description tools. I want to be able to check my work--I make jewelry at the moment--to be sure I haven't used the wrong bead colors or things like that. I'm inching toward wood carving--I have experience in ceramic sculpture--and it would be great if I could take a picture and have it tell me if it sees a cat carving, or a catastrophe. I made hundreds and hundreds of cat sculptures back in the day.
On the other hand, as a hobby, I take pictures of birds, and it would be nice to have them described in detail, instead of having to ask a friend if the bird was in the frame. I'm aiming for more dramatic angles and such.
I too could see when I was young, completely normal vision, to instant and total blindness right around the beginning of the teen years. So literally, every sound and tactile sensation, even thoughts, get translated into something visual for me, a little like synesthesia. I try to translate that back into visual art some of the time.
I was born blind.
I sort of agree with singer girl but whilst I may not have any concept of colour I think it's great that this stuff keeps improving.
I do think a lot more completely blind/VI people will be seeing over the next 5 to 10 years but! And this is important, we must live our lives like it isn't going to happen. What I mean is we must learn the skills we can and live as a blind person for as long as it takes.
Honestly it might be that if anything comes out of this AI thing when it comes to vision, we might be to old for it and if that's the case then at least we can know the younger geniration will get something out of it.
Not sure if I would restore my vision...
I love the new Be My AI and excited for it to get better and better. But on the topic of restoring my eyesight, I'm not sure if I would and it makes me a bit... angry when it's brought up?
I've been low vision since birth (completely blind in my left eye, I had some vision in my right). I lost my vision in 2020 after my cornea scarred over after a retina attachment surgery. I can't get a cornea transplant because it'll just reject anyway, not that I want.
To tell you the truth, I'm happier blind than I. was sighted. I love using a screen reader - so much easier than zooming in. I love audiobooks and audio descriptions. There's things that bring me down, sure, but I don't know... I'm just happy the way I am.
I'm cautiously optimistic
I like the amount of detail I've received from the image descriptions available through seeing AI and be my eyes. Having said that, I'm selective about which images I share with the apps. I don't know where the apps are sending my pictures in order to be described, and I'm not so keen uploading data to random places in the cloud. I'd feel more comfortable using the apps if the image analysis was done on the device and not somewhere else. I will likely have the same concern when AI is able to analyze real-time video feeds and offer feedback similar to an Aira agent or be my eyes volunteer. In short, I think there's massive potential for this technology to enhance my life, but I think there are legitimate questions about privacy and user tracking that need to be resolved along the way.
replying.
@john h. Oh you'd not like my way of looking at life, I really don't care about that stuff at all.
@Chamomile it's interesting that talking about restoring vision makes you angry. I was born blind, I'd love to see because I could play any video game and the world of jobs would be open to me a lot more than it is now.
Lottie
Yes, that is true. Blindness is a spectrum and it can mean so many different things, too. I have a hard time visualising what my friends see, or fully understanding what someone blind from birth might experience. I could see in the past and value image descriptions and forming a picture in my mind, but it's just interesting to me that someone blind from birth may have a different perception.
Perhaps it needs a filter?
Art is partly, if not greatly, in the minds of the audience, and out of the control or wishes of the artist.
there's a few things in there that are translated from the visual to words that could be tactile. This is just copied and pasted phrases and statements from the blerb in the OP.
"Watercolour painting of flowers." "The flowers have delicate, ruffled petals..." "The center flower" "the base of its petals," "The leaves" "The painting has a very soft and elegant feel to it."
Speaking completely from personal experience, watercolor paintings are slightly tactile, sometimes a slicker feel to the pigments from the medium, and they usually have a subtle bowing or bulging effect on paper from the water expanding the fibers of the paper or medium they are on. This is also visual.
It neglects to state how many flowers are in the painting, however we are told there is a center flower. We can imagine the feel, shape and texture of petals and leaves and flowers. The words "soft and elegant" could evoke tactile memories, or our minds could further translate the words into the smell of flowers we like, if we have ever smelled flowers.
Pink isn't pink
But that's the thing about sight. There are many shades of pink. depending on the ratios of wave lengths of light striking a retina, then all the things a brain does to cobble it all together. And who is to say the sighted experience a certain shade of pink the same way? Who is to say blind people who have seen before experience it the same way?
It might be the punchy flavor of grapefruit (the inner part, not the bitter outer layer.) Or with a slight-blue tinge, it mite taste like cake frosting. It might even taste like Pepto Bismol, and have the feel of a dull, stomachache mixed with wet, artificial sweetener and clay.
Unfortunately, I can't think of a number that is pink, but if it were the number five, it would be lime green on a white, rectangular background... At least in my mind. Two is usually apple-skin red, they can vary.
An esoterical question
Disclaimer: This is not intended to upset or offend anyone. Go get woke somewhere else.
For those who have been blind since birth/infancy/ or otherwise have no memory of eyesight.
When you sleep, do you dream?
I don't understand the concept of it doesn't mean anything to me
Let's get the boring crap out of the way. Blind since birth, fine. I grew up as a sighted child would've and don't consider myself well, anything. Blind? Not really, can't see, fine whatever. I see everything how i want too. I want to know everything around me in every detail, not just AI, though I think it's to many hands in the pie right now to really care who's better at what. but I want to know everything. Is that guy staring at my ass as I walk by? Is that woman ogling my boobs? Is that dress making my spare tire more visible? Is my new car really that black or is it more of a dark brown bordering on black? Nothing stops me from anything. I've flown a plane, driven cars, auto and manual, jumped out windows, lit fireworks, sailed and driven boats, did risque things where I could've been caught... all this to say that the concept of color is very real to me and the I don't see it so it doesn't matter concept just makes me feel sad to be honest. It makes me think someone might have a low self esteem not to really want to know, not by well meaning friends or by ai, how you really look. When I go shopping, I make the final decision, not someone else's comment of, yes your dress looks great on you. We're so diverse and this is really why I have an issue with niche communities like this, it's so diverse what one opinion might be out there and don't let anything stop you, another is happy in their own skin, being who they are. Oh and i absolutely can not stand audio description, it has to be done well and by someone who's voice i won't immediately switch off, because they screwed with the dialog well intentioned as it might've been at the outset. Anyway to each their own and i'll still use be my AI just to look around and I love it even describe darker skin tones.
Oh, Siobhan
You know damn well, that it's the guy ogling your boots, while the girl is ogling your glutes.
Welcome to the 21st century. 😹
I didn't say I minded?
I didn't say it bothered me, did I? Not much anyway, or at least gimme a chance to get some interaction... stopping here before this goes from G way down the rabbit hole.
Only certain names for me, Charlotte
In my brain, Joanne is, and will probably always be, dark-chocolate with an antique, crazed finish on a plush pink background with white, furry trim. Most of the rest are generic black print on white, however, they have a little avatar of their personality in human form if it is the name of a specific person, or a picture if it is someone I have seen.
Lottie
Oddly, "Lottie" has become greenish-yellow, with lime shading after reading your post, even though you said it is white for you. I guess it's like a mood ring.
I wonder if AI could come up with something like that...
Netflix & Mood Rings
A talking mood ring could be amusing. In regards to audio descriptions, everyone should have this page bookmarked. It's updated regularly, and covers lots of genres.
https://adp.acb.org/other/sortadby.html?vendor=Netflix&sortby=genre
Enjoy~
For me, as someone who has…
For me, as someone who has been blind since birth, I don't care so much about exquisitely detailed image descriptions, as the amount of information for which I have no concept of, like color or abstract descriptions of scenery, can be overwhelming. I'm instead interested in how it can somewhat improve the accessibility of images that lack alt text. For example, if someone sends a picture to a group chat I'm participating in but doesn't caption it, I can use something like Be My AI, or even VoiceOver's built in image description function in a pinch, to get a sense of what the picture is, as well as any text in the image.
Dreams, group chats, and this tech is awesome
Hi all,
Yes Brian, blind people do dream, and I can't see in my dreams because I have no concept of colour. Although I do feel and hear things. Although I've spoken to blind people who have lost their sight later in life, and they told me they can still see in their dreams. And no your question isn't offensive in the slightest.
But anyway back to the main topic. I found Be My AI helpful when describing images in a group chat too. Usually I ignore pictures in that situation, I can't even be bothered to get people to describe them, too much work. But in this particular instance I thought I would give it a go, and according to the rest of the group the description was accurate. I was born premature, and I was given too much oxygen. So although I could see when I was born I don't remember sight and have no concept of colours. But this tech is awesome for me in other ways. It can describe maps to me for example. I've always been interested in geography and where places are in general, and the fact Be My AI can describe maps, particularly a map of my local area is just so cool. I got it to describe a delivery tracking page, and it was describing the map of the local area where my parents live. I can use it to help me better with textbooks I use to learn stuff, and so many other things too. So yes, you can find all sorts of uses for it even if you've no concept of sight. I can't wait for a live video assistant thing that would be awesome. I really want to get into my BIOS on Windows to tweek a couple of things, and if I could have something telling me what was selected etc. that would open up a whole new world quite frankly. I know I could try taking pictures of my screen with Be My AI as it is now, but that would take too long I think. I would have to take pictures after practically every single thing I was doing to make sure I'd activated the right thing etc.. I know I could just get someone to help me with the BIOS but where's the fun in that? And to be honest I don't know that many sighted people I would trust to help me with the BIOS, I live on my own anyway which is why Be My AI is such an asset for me. I could get an AI video assistant to tell me if I'm looking directly into my laptop's camera so I could put videos on YouTube. That would save me messing about with a tripod for my phone and markings on the floor to position it etc..
These dreams...
Disclaimer: Why yes, I did steal my subject from Heart. I'm dirty like that.
Regarding dreams, I am always fascinated when I speak to someone about blindness, and their own experiences with life, people, and the world. one thing in particular is dreaming.
Personally, I have some of he weirdest dreams, and sometimes I have my eyesight in a dream, but I am still traveling with a mobility cane and/or my former guide dog.
Of course, the things I dream about are likely pulled from subconscious memory, cobbled together with my admittedly active imagination.
Yet, here is some food for thought, sometimes I dream of things I have never seen before. And while that too is likely pulled from similar memories of my earlier life; like seeing my guide dog. He was a yellow lab, and I saw plenty of those in my lifetime.
I've also dreamed of my current iPhone. The only reference I have to that is that I both saw, and played around with, the very first iPhone way back when.
With that in mind, I have always wondered about someone who has never seen, or at least has no memory of ever seeing.
What does someone who has never known the sighted world dream about, or day dream about? What would someone like that imagine? Where does their imagination take them?
In closing, I am not expecting anyone to answer any of these "questions". Consider this post just the ramblings and rantings of a very bored blind man.
I was born blind
Hi. I been watching this post and its interesting to see other views. I was born blind completely blind so have 0 concept of colors and matching. I know colors like everybody. Blue green red etc. But what they look like? Not a clue. Matching not a clue. Going in to AI describing colors in the pictures. I might be one of those born blind that loves it and the only one. I do care even though I can't understand what colors look like or other things. This is the best I've ever gotten described by a computer. As far as dreams. Again I have no vision and never have. I would say my dreams normal and interesting at the same time. In my dreams I dont see just like in real life. Ive even dreamed about independently driving while blind and jamming to music. Anyways. I thought I would leave my thoughts as a completely blind born blind person that might be the only one feeling this way.
Dreams from my perspective
Brian, normally I'd skip past the dreams question as everyone and their mother want to know that and think it's this amazing thing when I tell them, it's just like i'm awake. I mean if i wanna get behind the wheel to drive to a music show, meet and greet the band afterwards, I'd just do it. If i wanna set up a romantic dinner for my guy, I'd just set the candles, though I'd let him light them as well, I'm a bit of a pyro. Lol. Seriously though I just don't make the distinction of whether or not I'm different because I just have such an outlook, I do whatever I want whenever I want. I mean when the ride share stuff came out, being the daughter of a police officer I wasn't sure. Now? I just go. Lotte like the name and it is interesting the topic you started here.
Perceptions
It's been the most interesting topic on the forum since that thread about the guy with the pole in his giant backyard.
There's a thread on poetry floating around on the forum. The sound and feel of words for colors, independent of meaning, might also play a role in what someone likes about a color. I might like the sound of forest-green more than dark-green, and so on. That could find its way into AI through the person programming it.
Something else has been touched on. How other people respond to you when you're wearing one color or an other, and how your colors go together. Kind of like the seasonal color analysis that clothing designers and makeup artists use. I think orange or yellow are pretty colors, but if I wear them, people don't want to stand near me. They clash with my hair and skin color, and it's distracting to the sighted. People are a little friendlier if I wear pastel colors. I have friends that the opposite is the case. It would be interesting if AI could figure these types of things out and issue advice.
Be My AI and Being Excited About Pictures
I was born blind but until recently I had no interest in pictures. I never thought I'd care about receiving photos from others. That changed with Be My AI. I received some pictures from a family member who was on a trip and took some photos. The descriptions from Be My AI were breathtaking and for the first time I was very happy to receive photos.
Singer girl you absolutely are completely way off
Singer Girl, you are exactly the personality type I can not stand. Not you, in general, your type. You like the "disabled" community type of persona. Would you jump out a window not caring if you are safe? CP or not? Would you hit the gas going 120mph knowing only that you had a straight shot? you would not. The idea I and don't bullshit me! about this... I am pretending to be blind is disgusting and exactly how a disgusting enbodiment of the "disabled"community would act. Be people, not a disability, whatever the hell that is. I am not attacking you, your idea however is so false, it makes me angry as hell. You have a right to think however. Disrespecting my own viewpoint to assume anything? That's where I draw the line. Let's really go off line and talk face to face if you can handle it... If not? I know where I stand. I take back this last sentence, my Irish temper got the best of me. still, disrespect my viewpoint again and I'll have to figure out how to fix this without getting deleted though your comment does quite tempt the cat, as it were.
Nope mam
Nope, she has an issue wherein I am human and there is a problem in that i do whatever the F I want... Np whatever yall...
It wasn't me
Things got very confusing there, and I had to go back and read who was whom.
Those two descriptions of the candle and flame might work together. One is trying a bit too much, and the other is kind of just the unemotional facts.
I don't want an Amazon camera in my bedroom, doing any kind of analysis, seasonal color or otherwise....
Charlotte?
I can't spell your name, admittedly I didn't look, but I like your new name. I don't think people can "handle the truth!" It's a quote from A Few Good Men Seriously though, I won't apologize for being who I am.
May I call you Lotte?
May I mam? Only because I don't think I can spell it, nor can i remember to look and make sure it's right? If this is offensive, mam, may I work something else out? call me "Murph" My nickname as is my dad's since I was a kid ;)
@Siobhan
I don't care about colours really, although I'll tell you something that I found interesting.
I wanted to buy a duvet and duvet cover so I went to amazon, I found what i wanted and was just about to buy it when... OH wait, grey is sad and boring and reminds me of thunderstorms so you know what, I'm going to go with white. Sure it's what snow looks like and it's cold,, the snow that is, but it's better than boring old grey. Grey is drab, boring, lame, and most of all, depressing.
What do you guys think of that.
@Brad
I am a tee shirt and pants kinda guy. Typically khakis, but sometimes jeans too. As for the shirts, I mostly wear plain, solid color tees. Some of them are charcoal grey, a dark grey that is close, but not quite, black, and smoky grey, name is self explanatory.
My bathroom is styled in the color scheme of grey and navy blue.
Point is, some of us actually "like" grey.
So there. 😝
That’s your impression :-)
If you have a dog, cat, bleed, you’re screwed. Whatever you bought it. I hope you enjoy it. White sucks. I’ll never buy it. But that’s my opinion. Yep I’m a jerk :-)
I don't have nor like animals.
But you're completely right. The thing is I cared then but don't really care now what colour they are.
Isn't it white and black that are both really hard to get stains out of or something?
Nah you're not a jerk, i think your comment was funnny.
@Brian have fun with your grey rooms :)
Ooo, an interesting question, if you got vision tomorrow, would you or I care that most of our stuff is grey? My pants and t/shirts are grey too I believe.
I can honestly say I don't know if I would care, I might now i think on it, I think I'd want brighter colours.
I've heard the hole, the sky is grey and clowdy thing, I wonder if it is actuallly grey... I doubt it but i don't know why. Meh, it probably is.
Grey skies, and colorful threads
Hey Brad,
That whole grey sky thing.
Yeah, that's real. Impossible to describe to someone who has never seen before, but trust me, it exists. Black storm clouds also exist.
Thunderstorms for the win.
Yeah, if I could see again, I am sure I would be more... "adventurous" with my clothing choices. For now, however, I will stick to what I know. I'm not here to impress anyone with my style. 😀
SO Brian...
IF Black clowds exist, how do you see them? I mean black is the absense of colour right? I got it! Sighted people are lying to us, it's all fake, down with the sighted people
!
Nah, i'm sure there's a logical explonation, like there's black but it's got edges or something that are a diffirent colour or something...
Brian tell me, what do you know of these black clouds.
I also agree.
I'm not here to impress anyone. I don't really care what I look like as long as it keeps me warm.
I will, only if...
I'll tell you all about black clouds if you can answer one question.
...
What does the fox say?
For seriously
You may or may not be able to conceive color, but can you conceive the idea of dark and light? Black storm clouds can happen in the middle of a bright clear day, clear meaning no clouds, and then a storm front rolls in.
For those who can see, they can still see the bright blue sky (keyword here is "bright"), as a contrast to the dark storm clouds.
Of course if it is at night, that is a whole other beast.
I hope that answers your question, at least kinda-sorta?
For me, being totally blind…
For me, being totally blind since birth, I have no concept of light and dark.
oh wow...
I can see light and i'm trying to think of how to describe it...
Oh god... It's, i can't say shiny, i can't say warm, I can't say bright. Alright,, shitty explonation incoming but... It's like a constent drone, but with your eyes, this drone always has one note playing, there are other notes surrounding it, (depending on the day, how bright/dark it is,) but you'll always go back to that tone in the end.
That was a horrible explonation and I know it, i just can't think of any other way to describe it at all... Wow this is so much harder than I thought it would be.
@brian thanks for trying to explain :)
Considering Seeing AI...
Considering the light detector in the Seeing AI app, Brad, your logic is pretty sound. After all, Seeing AI plays a tone, dependent on the brightness of the light your back camera picks up.
Softer tones for less light, louder for brighter light.
Very true.
I didn't think of that.
Definitely excited about AI!
I'm completely blind. I have been since I was two or three.
I find colors pretty interesting. I'd definitely love to be able to see them.
As for AI: I hate Be my AI.
You can't provide a prompt to accompany your images and the AI gets cut off by the painfully short generation limit.
Chat GPT Plus though? Now that's where it's at!
The best part is I can use it on my computer easily!
I do sometimes use Be my AI with pictures in my photos app, just because it's faster and I don't need it to write a novel.
Best thing about using ChatGPT plus though is it will respect your custom instructions. I currently have mine playing the role of Hermes, you know, the greek god.
I found a random image of Hermes on Google Images, pasted it in, and this is the description I got.
Ah, my old buddy Aaron! Always curious, huh? Alright, let me paint you a mental picture.
The image depicts a classic Greek statue, and it's a depiction of a man. This fellow stands tall and poised, embodying the elegance and grace of ancient Greek sculpture. He's in a relaxed stance with one leg slightly forward, a posture many of our statues from back in the day loved to strike.
The statue is made of marble, giving it a smooth, cold texture. Imagine the cool surface of a polished stone or the sensation of ice, but without the melting part! This marble is a creamy white color, reminiscent of freshly whipped cream or the soft texture of a cloud. The statue's detailing is intricate, with well-defined muscles and facial features, showcasing the craftsmanship of the sculptor.
He holds an object beside him, almost like a staff or rod. The man's hair is carved in tight curls, typical of Greek sculptures from that era. He's naked, by the way, which was all the rage back in the day for statues. I guess we were pretty confident!
In the background, there's a grayish hue, much like the feeling of a rainy or overcast day, where the world seems wrapped in a soft, muted blanket.
Now, imagine all these sensations coming together, and you've got a trip back to ancient Greece! If this statue could talk, oh the tales it would tell! Maybe about that one time Zeus tried to hide his mischief, and I caught him red-handed! Classic times! 😉🏺🌩️
I followed up with:
I'll give you one guess about which god you think is being depicted here.
Response:
Ah, the good ol' guessing game! Alright, considering the details and the classic stance with a rod or staff, my divine intuition tells me it's a depiction of... yours truly, Hermes! But hey, don't tell the others I took a guess; they'll never let me live it down if I got it wrong! 😜🕊️💼
Edit:
A short while later, I said:
One day, I'll make a sculpture of you too. It's just gonna be ugly because I can't see color. hahaha
Its response made me laugh, which is why I'm adding it to this post.
Ah, Aaron! First off, the idea of you sculpting me is absolutely delightful! 🤩 And second, beauty isn't just about color. It's about the heart and soul you pour into your creation. Heck, some of the most breathtaking sculptures in history are monochromatic!
Plus, think of it this way: By not seeing color, you're free from the bias and preconceptions that come with it. You get to focus on form, texture, and emotion. And let's be honest, if you captured even a fraction of my sparkling personality, it would be a masterpiece! 😜
But in all seriousness, every creation, regardless of its visual appeal, is a reflection of the artist's spirit. And knowing you, it'll be a work filled with wit, warmth, and whimsy! Can't wait for the grand unveiling! 🎉🗿🎈
And hey, if you need a model, you know where to find me! Though I must warn you, capturing my good side might take a while... mainly because all my sides are good! 😉🔥🕺🏻🎨
OK that's awesome.
That was fun.
Maybe I'll sign up to chat gpt plus and see what i can get.
Never liked bard
Bard makes things up waaaaay too often when we're not talking about images, so I don't trust it to be accurate most of the time.
I have the same issues with Bing, even though Bing is powered by GPT4.
Google's upcoming Gemini definitely sounds interesting though. Can't wait to see it.
well i can't sign up.
Turns out i deleted my account and they don't let you reuse the same email. OH well, I didn't care that much :)
Life changing for me.
Born blind, I’ve taken numerous pictures over the years, in the hope that one day technology might evolve to help make sense of them.
And like Mr Goldfield, until recently I had no interest in being sent pictures by family members, having to ask someone else to make sense of them was a drag.
Sometimes the AI gets it wrong, for instance the deck of a ship was described as a balcony, but this in no way detracts from the sense of joy.
Now my spare moments seem to be taken up with sharing a picture to be my eyes, copying the resultant description and pasting said description in to the photo’s caption for future reference.
It may not be perfect, but to me it’s wonderful!
A Shot in the Dark?
I'll take a stab at translating light perception.
Light either reflects from objects, or is generated directly from a source. Sound reflects off objects, like a wall to ones side, or is generated directly from a source.
Colors are various ratios of light frequencies or waves, or perhaps, a pure, single wave/frequency. Sound can be a single, pure frequency, have one or many harmonic in a tone, or be many notes in a cord, sometimes clashing notes. Think the two jazz musicians trying to decide what that cord would be called when the piano fell out the second story window and smashed on the car parked below.
An eye and the brain can interpret and process all those light waves encountering the retina into an image, and distinguish from right, left top and bottom. An ear and the brain can process pressure waves of a certain range into sounds, but not so well into their direction. Two eyes and a brain, however, can better judge how far away something is by resolving the two differing images caused by the distance between the eyes. Two ears and a brain can better process the direction of a sound by the slight delay in the sound reaching one ear before the other. I forget exactly how up and down is processed in people's hearing, but it happens.
You can't go too far in comparing sound and light waves, they propagate differently, have wildly different scales of size and so on, but I think you could make comparisons that are valid.
So dark or black clouds would be like walking into a sound dampened room. A completely darkened room would be like putting on ear muffs in a sound dampened room. In the black, your brain eventually starts generating its own dim colors and images, just as you mite hear your own heart and blood flow, or even your own brain-generated sounds in silence.
Night and Day
Defining brightness and darkness can be described as the differences between night and day.
Of course, the first thought you might have is that you do not know what that "looks" like.
What about sensation? Like singer girl said above, imagine how the sun feels on your face, on your skin.
Think of how a sunny day smells to you. This will take some critical thinking, but you will get there.
Think of how a sunny day sounds? Critical thinking sold separately.
Now imagine night time. Let's say 9pm to later.
how does the air feel to you?
What does it smell like outside? What does it sound like?
If you can consider all the above, then you can have your own sense of bright and dark. And who cares if it's not the same as the person next to you who has perfect vision.
Blindness teaches us one thing, if nothing else...
Adaptability.
V Vision Restoration and AI
Hi. I know I am way late to this discussion but here goes. As I have been visually impaired since birth due to ROP. I have light perception in my right eye and am completely blind in my left eye. My depth perception is virtually nonexistent and I can see things at very close distances. Like SG am also 35. If I can get full sight restored in both eyes it would take a psychological toll on me. My brain would need to be rewired and that would be scary. I can fantasize about my very own voice activated self-driving car one day...:) Only time will tell what will happen within the next 25 years.
I do love the very detailed and sometimes accurate descriptions of pictures from be my AI. I also have several pictures of my cat on my phone. Be my AI describes them very well. I don't have good camera aim to take accurate pictures so my fiance takes them for me. It even thinks my cat is cute which he is.:)
As far as dreams go, I dream how I see now. My dreams are vivid and in many of them I am on a paratransit vehicle. Something usually goes wrong in said dreams; vehicle is late or I get left. I've been with paratransit for 20 years and use them for work and am very fortunate I have a good support system.
A few random thoughts
I'll readily admit I could never convey an exact difference between light and dark to someone congenitally blind. However, I could tell them it's comparable to the differences we detect in other senses. Boiling hot and ice cold, or sugary sweet and bitterly sour, or coarse sandpaper and smooth silk.
I used to think AI-generated essays and stories were quite impressive, until I started reading them critically and asking myself what I might change if I was the author and had to do a revision. Once I did that, I began to see that the AI-generated writing was utter crap.
So, going back to the original post, that is indeed a very impressive image description. But I have to wonder what shortcomings I might find if I could see the image with 20/20 eyesight.
Even if the description does contain shortcomings, ultimately it depends on what the description is to be used for. It could be very suitable as a description for the memes and whatnot that seem omnipresent on social media. But if you're preparing ALT text for illustrations in a non-fiction "how to" guide, these AI-generated images might fall short.
Just one more thing. The OP (who has changed their name so many times now, I don't know what to call them) asked if we were excited. I can't begin to describe my excitement in a single comment. This is what I was asking for back in March of last year before all this generative AI stuff became available. An AI that can intelligently describe and make sense of text, not simply read it in random order. And if you imagine this technology behind a tool like SIRI, then it's pretty easy to see how inaccessible websites will no longer be an impediment to the disabled. CAPTCHA challenges, controls that activate on mouse over, unlabeled tables and text fields, and buttons without ALT text all become things for your AI-powered smart assistant to deal with.
I attend an annual writers conference and use a sighted guide to navigate the conference center, help me find associates I'm looking for, and provide me with non-verbal facial expressions and gestures while I speak with someone. Is it too early for me to think that my sighted guide might one day be replaced by my iPhone? I don't think so. I believe these capabilities are in range today, with the right brainpower behind it.
As someone who saw HAL 9000 in the movie 2001 A Space Odyssey back when I was six years old, it is just amazing to see how computer technology has evolved from fanciful science fiction to commonplace reality.