#image description tutorial

[rjalker]

if you see a screenshot of a tweet that has ALT text, before you reblog thinking "Oh, great, it's already described!" check the ALT text.

If the tweet includes a picture, the picture is probably not included in the ALT text.

When you share a link to a tweet (as of December 7th 2023, who knows if/when Elon Musk is gonna break this feature), it automatically converts itself into a "screenshot", with ALT text automatically generated listing the poster and the text content.

It does not do anything for any images in the tweet, because it's literally just copying the text.

You have to manually describe the images that were in the original tweet.

And because most people don't need image descriptions, they will just see that the image has ALT text, not check to make sure it's accurate, and reblog without realizing they're reblogging an inaccessible post.

Please check that ALT text is correct before reblogging. Even if it's not automatically generated, some people are just using it completely wrong because they don't know what it's for.

If the ALT text is incorrect, create a plain text image description, explain the problem, and ask the OP to add the full ID to the original post.

[describe-things]

Hello do you have any guides/tips for writing image descriptions, I’m always at a loss of how much/little to include

Here's a post I made a bit ago going over the basics for what and what not to do.

Here's a tutorial showing different levels of detail.

I usually will include first, what is the medium? A photo? Digital painting? A traditional painting? (in which case I'll usually just say 'painting')

Here's another little tutorial for this post, showing images of increasing detail, and descriptions to match, which will hopefully help.

The main thing is to make sure you describe everything that's important in the image.

A good way to learn is to follow a lot of blogs that do image descriptions, so you can see how different people describe different things.

A new tutorial for April 2024:

simplest:

[ID: An MS Paint drawing of a grey square on a blank white background, saying, "Hello from Flatland". There is a simple black border around the image. End ID.]

a bit more detail:

Note that there's nothing inherent in the image to tell you this character's gender, but because I, the describer know it, I can simply use the correct pronouns and convey it that way.

If you're describing people whose genders you don't know, it's best not to assume.

[ID: An MS Paint drawing of a grey square on a blank white background, saying, "Hello from Flatland". He has a simple eye drawn on the upper right corner where his dialogue is, and has a pink heart and a brain represented by a circle with squiggle lines inside his body. There is a simple black border around the image. End ID.]

Some more detail:

[ID: An MS Paint drawing of a square on a blank white background, saying, "Hello from Flatland". He has a grey outer layer, with his insides purple, with a small eye just inside an opening on the corner where his dialogue is, with the corners appearing almost like a beak. He has a pink heart and a brain represented by a circle with squiggle lines inside his body. There is a simple black border around the image. End ID.]

Yet more detail:

[ID: An MS Paint drawing of a square against a blank white background, saying, "Hello from Flatland". He has a grey outer layer, with purple insides, and various blob-shaped internal organs in different shades of pink and purple to reddish-purple, with all of them general blob shapes except the purple heart, which is in a simple heart shape. There is a simple black border around the image. End ID.]

Ooh look at this:

[ID: A digital drawing A green sphere, which says in matching green text, "Hello from Spaceland". The sphere and its dialogue both cast shadows on the white ground below, where a square, tilted at an angle as though lying flat on the ground while we view him from an angle, says "Hello from Flatland". The square has a grey outer layer, with purple insides, and various blob-shaped internal organs in different shades of pink and purple to reddish-purple, with all of them general blob shapes except the purple heart, which is in a simple heart shape. End ID.]

I'm too lazy at the moment to twist my brain to draw a tesseract so you'll all have to use your imagination for the image that would come next lol...

Hopefully this helped demonstrate a bit!

[jackgoodfellow]

Anyway, it turns out setting art goals according to what Julia from Drawfee can do is like setting swim goals according to what sea otters can do.

[necromancer-mango]

[start image description: Digital drawings of various poses on multiple pages. There is some text on the pages. In the first page the text reads: "Spend a lot of time looking at how people stand. Observe yourself to avoid stiffness, curve the body the body/shift hips and posture of the spine. Arms can break up lines." The second page's text reads: "More focused. More open/relaxed. Very relaxed/sleepy. Foot skids in, tilt for speed. Changes to spacing of leg without changing anything else alters feeling. Consider what you want to convey through posture."  The third page’s text reads: "Firm pose, weight on this leg. Same pose but body is angled. Less firm, looks cocky. Weight on this leg. Not very bendy (body straight), but angle gives attitude. Angle body plus adjust perspective can make it pop. Focus is on the upper body (especially arms and face)." The fourth page’s text reads: "Big curves, big movement, big emotion. If something is mostly straight, pop something out so it doesn't look stiff. Arm pops out, spine and leg. Looks down, can be calm or uncertain. Body droops, pops out. Weight." The fifth page has no text, rather there are multiple sitting poses on a page. The sixth page’s text reads: "Dramatic poses can be derived from dance, sports, fashion poses, et cetera. Leg pops up, push down on this leg. Extend! pops out. Weight still here." The seventh page’s text reads: "Position draws attention to certain body parts. You want to guide the eye. Foreshortening can be a curve. Interesting bits: head and hands position (torso is the rigid part due to position)." The eighth page’s text reads: "Extra examples." /end image description]

Body poses! Trying to make this was actually kind of difficult. I struggled trying to articulate how I placed people in spaces considering quite a bit of it is intuitive at this point. This guide’s focus is less about anatomy and more about flow and bodies in various positions and amount of motion. Sometimes the most difficult ones are the relaxed ones, mostly because it is subtle and can quickly become stiff. Hopefully this can be of some use. I think another thing that might help with positioning the body is to work on conveying physicality of the person in the image (how the flesh squishes and rests in contact with other things). I also spend time looking at other art as well as other bodies. Looking at bodies (including yourself as a reference) can really be done anywhere in any circumstance. For example, I did 6 yrs of musical theater with 2 of those 6 years doing tap. Staring at other people plus yourself when learning and practicing how to dance increases your awareness of your body position with others. This awareness can be utilized in other things outside of dance, as can be seen with how I place bodies.

[pathos-logical]

hello! saw that you add image descriptions and I was wondering if you think the alt text feature tumblr has is good for image descriptions or if it’s better to just write it out in the post? I’m not the most knowledgeable on how to do good image descriptions but I wanna start making the habit of putting them on things

Hello!!! This is a great question, thank you for reaching out!!! The alt feature is great for descriptions, but the most recent consensus I've seen is that doing both alt text and plain text (body text) IDs in the post is actually the most broadly accessible way to describe a post!! Here's an example of what I mean (you'll notice there's some more detail in the plain text, and that the alt isn't appended by brackets or extra ID formatting):

[ID: My icon, an emoji fish carrying an emoji love letter in its mouth surrounded by hearts and set against a pink gradient. End ID]

The blind community has competing access needs, which means there isn't a one-size-fits-all format, and your final choice comes down to what you prefer on your blog, as long as your IDs are useful and clear :)

Some bullet points to help you get your bearings:

Alt text is occasionally plagued by glitches and is sometimes considered less accessible to people who may not use a screenreader. It's still the most useful form of ID for screenreader users, from what I've gathered!

Plain text IDs should be written in fully plain text-- no small text or excessive italics/caps lock! However, I've seen that bold is generally more acceptable as special formatting. Utilize line breaks if necessary, but try to emphasize clarity and conciseness rather than exhaustive detail. Also, IDs go directly below (or above) the image, not at the end of a post!

People who do use both forms of description will frequently, as I did above, write a relatively bare-bones alt and a more in-depth ID so screenreader users can choose whether they want to sit through repeated information or not!

And last: Thank you again for reaching out!!!! Every person who chooses to make their spaces accessible makes the world a better place :))))

[smilepilled]

⠀

【🪲】 Hello, hello! Here, I will share some questions I ask myself when making image descriptions. Make sure to remember that these are personal questions, and that they may not work for everyone — remember to find your own way to work with image descriptions, as to be accessible for yourself too. 【✉️】

⠀

What is this? — A divider, banner or icon? Maybe a replycon? A flag? A drawing, an animation? A graphic element? Is it commonly referred by with a specific name/term? Is it decorative or informative?

How is this? — Is it round? Is it square? Does it have a specific geometrical feel to it? Is it sharp? Is it soft? Is it a multitude of shapes? What vibe do these shapes give me? Is it classy? Is it cyberpop? Is it minimalistic?

What colours are here? — Multiple shades of pink and yellow? Which && what names? Vanilla cream? Burnt yellow? Hot magenta-ish pink? Is it light(est/er/ish) or dark(est/er/ish)? Is it vibrant, neon or bright? Is it pastel, or a baby tone? What objects reminds me of this colour? Is it a gradient from (COLOR A) to (COLOUR B)?

How many details? — Is there additional images? Is there a symbol for the flag? Are there other shapes in the image? Does it have any outlines or stripes? Is thetr a gradient? Is there anything overlayed on it?

Who's this? — What character is this, from what series/media? What is their expression, and what may the expression possibly mean? What are they wearing? What is their stance? What are they doing?

Where's this? — Is it daytime or nighttime (if you can tell)? Is there a specific natural landscape? What does the landscape have, trees and flowers and some birds? How does the mountain in the distance look? Is it indoors? How does the room feel, scary or welcoming or home-like or official/stablishment-like? Does the room have any furniture? Which, and how do they look like? Is there any animal or critter in the image? What critter is it, and what does it look like?

⠀

【🌱】 Now, obviously, the amount of these questions being answer varies depending on how complex a image is — please remember to keep in mind that plain text must also be accessible to people who are reading it — a lengthy, continuous huge wall of text may give headaches and confuse people. This is also not a extensive list, and additions in the notes are more than welcome. Make sure to only answer anything that gives key information, if you'd prefer to keep it short and sweet... Toodle-o! 【🤍】

⠀

[cepheusgalaxy]

Y'all.

That post description post was right. So right.

Don't let those blogs do all the work. Describing just your own posts is not very hard, but when you try to add descriptions to more posts it gets really hard and a little tiring.

Try to make an ID whenever you remember. It can become an habit. Let's make images acessible y'all.

And if your blog has a lot of screenshots of like, tumblr and twitter posts, it's even easier. Specially if the original post already had an ALT description for whatever images it had in it.

Try.

[beryllineart]

How2Draw Comics: Conceptualize

So, it looks like the mouse adventure won the vote. Let's do this.

The first step to drawing and writing comics is to get a basic concept down on paper. For me, this comes in the form of creating my characters and giving them a world to live in. I also use this step to help me figure out what art style I'm going to use in this comic.

I usually figure out what art style I'm going to use before moving on to individual characters and the like. Above I figured out how to anthropomorphize a mouse and put expressions on its face. I also figured out how humans are going to look in this comic. Don't worry about individual characters at this point. Right now you are designing a world.

Once you know you can draw a generic character, it's time to look at making your characters unique individuals. One important thing about comics is not to make your characters too detailed. If you already have a visual concept that's pretty detailed, practice drawing them a whole bunch. As you draw, you'll find the less important parts of their design fading.

Draw them in different poses, outfits, interacting with friends, etc. Consider the world they live in and how that impacts their lives. You want to be able to recognize who is who. I gave the brother a nick in his ear, and the sister often carries around a spoon.

I assume you have a basic story in mind. This is where you experiment, by putting your characters in a bunch of different situations. I am an impatient person so I usually only do a page or so, but that page (plus my mind's eye) helps me figure out personalities, motivations, and lifestyle for my characters. Not names though. I suck at names.

Don't get discouraged if some of the things you draw look nothing like you imagined your character. This is why you practice! And who knows, one of your bad drawings might give you inspiration. Lady Giraffe was just a generic human design, but now she's... kind of the main antagonist. Yeah.

In summation, figure out your art style (how you draw your characters and settings), design your characters (who are they? What do they like, how do they react and interact, etc) and don't worry if your drawings don't look the best. The important thing here is practice. Draw your characters over and over again, and don't be surprised if they change a lot. I mean, my characters didn't, but I got lucky.

There are a lot of books and videos that will help you with character design if that's hard for you. They will do a more thorough job of teaching than I did here.

Next up is Story: writing the scenes and situations your characters will experience.

[terror-billie]

People aren't tagging Goncharov and the site is unusable! (or, how to filter post content)

First, hit your gear button and go to Account Settings.

In Account Settings, go to Content You See.

Here at the top is an option to filter tags, but scroll down and you will see Filtered Post Content.

This filters text found anywhere in a post, not just tags.

Add Goncharov and bam. Filtered.

[denzartriste]

Drawing bald people is a completely separate skill than drawing people, why can't i just give them hair why must head equal bald

[yeeiguess]

Disabled Yamaguchi art :)

ID : The first drawing is of Yamaguchi Tadashi from Haikyuu. They're sitting in a wheelchair, one leg up on the seat. Their right elbow rest on the raised knee, while the opposite hand is set on the upper arm. They have green hair with long bangs, braided on the sides and attached behind their head with ribbons.

They're wearing a lace choker, a shoulder off, large sleeved crop top in an off white color, and a skirt with ruffles in the same color. They're wearing pale blue short underneath, brown leg warmers, black Mary janes shoes and an Apple Watch with a rainbow strap.

They're smiling, and have acne and dimples as well as a 'they/them' tattoo underneath their collarbone.

The second drawing is of Yamaguchi and Tsukishima. Yamaguchi has a light green cane, a long layered blue skirt with lace at the bottom of each layer, and a corset with a sun/moon on it. They're wearing a blue bandana with moons and suns on it, a few necklaces, and green ribbons on their shoulders. They have acne and freckles.

Tsukishima wears a purple hoodie with embroidery on it: swirls, the moon, some stars, and a small sun on the hood. He also has brown pants with a turquoise ombré at the bottom. He has headphones on. They're walking together and looking to the side. The background is made of squares of different colors.

[rainbowcoloreddays]

Process of adding a skull design to an old zip-up hoodie.

Materials, in no particular order:

--gray zip-up hoodie

--white t-shirt, black t-shirt, red t-shirt

--white embroidery thread and needle

--red sewing thread and needle

--fabric shears, regular scissors

--notebook paper and pencil

--fabric pins, pin cushion

--thin-tipped markers

Step 1: I drew out my design on paper. I went through a couple iterations, and I even ended up changing my mind and attaching the horns separate from the skull. Also changing other parts of the design, like making one eye an X and not making a nose hole.

Step 2: I forgot to take pictures of the pattern I made but essentially I traced a salad bowl for the round part and freehanded the cheekbones and teeth. I only needed to make half of the skull pattern and I flipped it to trace the other half. I used pins to hold the paper in place and traced the pattern with a thin-tipped marker. When tracing fabric, it's easier to just make a dotted line with the marker. I learned this the hard way.

I then removed the patter and cut out the fabric with fabric shears. Since this fabric was from a t-shirt I first cut a chunk from the bottom before attaching the pattern.

Repeat for the eyes and horns. Fold your patterns in half when drawing them so they'll be symmetrical when you cut them out.

Step 3: Using white embroidery thread I attached the eyes to the skull. Pin these pieces into place before sewing.

Step 4: Using red thread, I attached the design to my hoodie. Pin in place before sewing.

And that's it's complete! I'm going to add plush horns to the hood and probably some plush spikes to the shoulders, so I'll come back with an update.

[silcoitus]

Tumblr alt text

Genuine Tumblr question: I noticed people include image descriptions in the body of a post (which, as an advocate for making the internet more accessible, I love!). Is there a reason people do that rather than adding the description into the alt text? (Which is pretty common practice for accessibility) is it just from not knowing how to? Or do people prefer that method more? Is it something I just don't know because I've only been on Tumblr since last June?

On the off chance it's from not knowing, you can add alt text to an image itself!

Add alt text to an image

I'll use this photo of my dog Yoshi as an example. This is for mobile, but the process is the same for desktop.

Upload the photo as normal.

Then, click the three dots in the bottom right corner. And click "add alt text."

Enter your alt text!

There are best practices with how to write alt text (see this article for some) but basically, include any text that shows up in the image and then a description of the focal points of the image. How would you describe this photo if you were telling someone about it over the phone?

If your screenshot has lots of text, you can copy/paste the text from where you originally took the screenshot. Or, you can use an app like Google Lens to pull the words out to paste.

If you want to check it worked, publish the post and look for the ALT icon at the bottom left of the photo. You can even read it by clicking it!

Hope that helps! Adding alt text is one easy way to be more inclusive with what you post online. Make it a habit and let more people enjoy your shitposts!

[necromancer-mango]

[start image description: Three pages of face drawings. The first page consists of a breakdown of the head and face, as well as multiple drawings of various faces. The text reads: "Eyes are located around the top of the ear, with the nose generally ending at the bottom part of the ear. If you place the eyes and nose first, you can build the rest of the face around it. This would be especially useful for certain head angles. The lower lids of the eyes will be visible as well as the nostrils. The area where the jaw and neck connect becomes more apparent. Lower ear sticks out. Push the face down further, and the lower jaw tucks out of view. Eyebrows and forehead can obscure the eyes. The lower lid of the eyes will not be seen well. The upper ear sticks out." The second page consists of multiple faces drawn at different angles, with a blue outline demonstrating how to break down the face into parts. The text reads: "The face is defined and oriented by the nose. Size of the nose and position should be kept in mind. Large angular noses can have a more defining nose bridge rather than a smoother connection to the forehead. Eyes placed closer to the nose bridge indicate a flatter face, there will be less shadows around the eyes. Fat around the neck makes the jawline look softer, and the neck thicker. There can also be double chins." The third page consists of multiple faces with varying facial expressions. The text reads: "Expressions! Don't be afraid to do wrinkles. You can also tilt the head in certain ways to contribute to the emotion." /end image description]

Made another one of these for faces and heads, as well as a page of various expressions. I’ll probably make a separate one for hair. The expressions are mostly examples, so I might also do another thing going more in depth about putting them together as well.

[hijab-described]

[ID in alt text]

Overview of some topics when it comes to drawing characters who are burn survivors.

DISCLAIMER. Please keep in mind that this is an introductory overview for drawing some burn scars and has a lot of generalizations in it, so not every “X is Z” statement will be true for Actual People. I'm calling this introductory because I hope to get people to actually do their own research before drawing disabled & visibly different characters rather than just making stuff up. Think of it as a starting point and take it with a grain of salt (especially if you have a very different art style from mine).

Talking about research and learning... don't make your burn survivor characters evil. Burn survivors are normal people and don't deserve to be constantly portrayed in such a way.

edit: apparently tum "queerest place on the internet" blr hates disabled people so much that this post got automatically filtered. cool!

[wind-and-cloud]

How To Make A Glamourous Fabric Fig Ornament For The Tree - Pillar Box Blue