AI 이미지 캡션 중 최고를 보여주세요!

Free, no account or log-in, perhaps not as good as the one you used :man_shrugging:

1개의 좋아요

For this game, you should be using the inbuilt Discourse ‘Caption with AI’ option in your composer preview that shows up when you hover over the image:

6개의 좋아요

Main challenge is to post only “the best” caption, will say that is this one:

This is a historical map titled “A general PLAN of the Town of Annapolis Royal in Nova Scotia,” surveyed by Captain Hamilton in 1753, depicting detailed fortifications, the town layout, surrounding landscape features, and a compass rose. (Captioned by AI)

Can help if this can translate handwriting to text in more than just captions.

4개의 좋아요

I don’t have the resources to play with the big kids, maybe you can caption for me :pensive:

1개의 좋아요

You should have that feature (it’s currently on TL2), though it looks like you used a link rather than uploading an image. When it gets automatically downloaded you can open it up in edit mode and see the option (or try it with an uploaded image).


Hmm. That still doesn’t seem to be an option even after it pulled it in locally. :thinking:

I think it’s because it lacks dimensions so doesn’t get the hover options in the preview to resize, etc.

![](upload://fCXxUNUVQkfUz1iTU48CnqKAFuf.jpeg)

If I add some made up ones then the options appear.

3개의 좋아요

A young boy stands with his arms crossed, looking upset, while a group of children in the background sits on a bench, seemingly excluding him from their conversation. (Captioned by AI)

4개의 좋아요

A child in a pink dress appears to be twirling or running playfully with a dog along a tree-lined path. (Captioned by AI)

My daughter dancing with our dog 10 years ago while we were out for a walk.
She is 17 now and our dog is 12.

8개의 좋아요

1개의 좋아요

An image created by AI and captioned by AI? :thinking:

An anthropomorphic duck dressed as a vintage pilot is seated in a military aircraft, with fighter planes and mountainous landscapes in the background. (Captioned by AI)

:duck: :airplane: :robot:

3개의 좋아요

I took your caption and generated an image from it

Here’s the AI caption for an AI generated image from an AI caption generated from an AI image:

An anthropomorphic duck, wearing a classic pilot outfit, confidently controls a fighter plane amidst a formation flying over mountainous terrain.

Surprisingly consistent.

8개의 좋아요

Error 500

:rofl: :man_facepalming: [1]


  1. Is it possible that error didn’t come from braindead content barriers of OpenAI, but format avif? ↩︎

4개의 좋아요

What type of files can I upload?
We currently support PNG (.png), JPEG (.jpeg and .jpg), WEBP (.webp), and non-animated GIF (.gif).

5개의 좋아요

A startled man recoils from a monstrous creature emerging from his laptop screen amidst digital chaos.

Now is caption describing image real good :joy:

2개의 좋아요

Although, not incorrect in it’s description, I think the AI has somewhat missed the point. Perhàps, I am expecting too much from the AI to get the nuances of expression depicting motherhood :woman_shrugging:t2::joy:.

2개의 좋아요
Side bumping this with…

Is there any proper survey of people with vision difficulty or using screenreaders, as to whether they feel this is a step forward? Perhaps comparing three cases

  • a community where the users have understood that it’s good to caption images
  • a community where the users are more average in their captioning
  • a community that’s used automatic captioning as seen here

What are captions for? Surely they are for accessibility. Is there a voice here discussing the accessibility angle? Discussing how the community culture is affected by introducing automatic captions?

This is a picturesque miniature view of a city with colorful buildings, palm trees, small vehicles, and a train running under a bridge that holds everything together <desc>. (Captioned by AI)

<English>At dusk, the setting sun illuminates the faded seed of a dandelion, which stands out like a glowing orb against the orange sky.</English> (Captioned by AI)

3개의 좋아요