PNG · JPG · WebP
or drop / paste an image
Your outfit breakdown will appear here
Upload a photo, an illustration, or character art — every garment, the hair, and the style, ready to copy.
Upload a photo and get every garment named — color, fabric, fit, and construction — plus the hair, the style it belongs to, and a paragraph ready to paste into a character profile.
Most people can see an outfit clearly and still not have the vocabulary for it — the difference between a trucker jacket and a chore coat, between washed indigo and slate, between a straight leg and a barrel leg. An AI outfit describer supplies that vocabulary. It is an outfit description generator that works from a picture: upload a photo, an illustration, or character art, and it goes through the look layer by layer (outerwear, top, bottom, footwear, then every accessory), names the aesthetic, and hands you a paragraph you can paste into a character profile or a story.
"Cropped ribbed tank", not "a top". "Oxblood leather Chelsea boots", not "shoes". The specific word is the whole point.
Describe 8 outfits a day without an account. Signed in, each description is 1 credit.
It never tries to identify who is in the photo, and it describes the clothing rather than the body wearing it.
Same four parts every time: the clothes one by one, then a paragraph you can paste.
Head to foot, one line per item: the specific garment name, the precise color, the fabric where it reads, the fit and silhouette, and the details that define it — seams, hardware, prints, distressing, layering. Outerwear, tops, bottoms, footwear, and every accessory from bags to eyewear to watches.
Color, length, cut, and how it is styled, including anything worn in it. A look is rarely just the clothes, and a character description usually needs the hair too.
The aesthetic named in recognized terms — streetwear, quiet luxury, dark academia, y2k, gorpcore, cottagecore — each justified from what is actually visible, plus the palette as a whole and the occasion the outfit reads for.
One flowing paragraph of 60 to 100 words, written as prose rather than as a form, so it can be pasted straight into a character profile, a story, or an image prompt.
From a photo to the words for it, in three steps
Drag in a PNG, JPG, or WebP — or paste from your clipboard. Photographs, illustrations, anime and game art, 3D renders, and fashion sketches all work.
The description can come back in any of 19 languages. There is nothing else to set.
Copy the breakdown for reference, take the closing paragraph for a character card, or send it straight to the image generator to draw the character wearing it.
Everything you need to know about describing outfits with AI.
Upload the picture and the describer reads the clothing straight away — no account, no button to hunt for. It names each garment specifically rather than by category (a cropped ribbed tank, not "a top"), gives the precise color, fabric, fit and construction details, then labels the overall aesthetic and finishes with a paragraph you can copy anywhere.
Yes. The style section names the aesthetic in the terms people actually search for, such as streetwear, quiet luxury, dark academia, y2k, gorpcore or cottagecore, and points to the pieces that put the outfit in that style. That makes it a quick aesthetic identifier when you know you like a look but not what it is called.
Yes, that is what the last section is for. Alongside the itemized breakdown, every result ends with a single flowing paragraph of 60 to 100 words written to paste straight into a Character.AI profile, another character card, a story, or an image prompt. It reads as prose, not as a form.
It does. Hair gets its own section covering color, length, cut and how it is styled, including anything worn in it — clips, scarves, headbands. A look is rarely just the garments, and character and costume descriptions usually need the hair too.
Yes. It works on photographs, illustrations, anime and game art, 3D renders, fashion sketches, and historical costume plates. On stylized art it will name the era, subculture or regional dress it draws on when those describe the look better than a contemporary fashion label.
It says so plainly, then describes any garments visible on their own — a flat lay, a mannequin, a rail of clothes, a folded stack. You still get the itemized breakdown; it simply is not attached to someone wearing it.
No, and it will not try, even for a recognizable public figure. It describes the clothing rather than the body wearing it, and stays with what is actually visible in the frame.
Without an account you can describe 8 outfits a day for free. Once you log in, each description costs 1 credit (refunded if it fails). Generating an image from the description is charged separately, because that runs an image model.
Need the whole picture rather than the wardrobe? Detailed scene descriptions, object lists, OCR, and custom questions.
Take the description and draw it — turn a character's look into a portrait with any of our image models.
Want the outfit as a generation prompt instead of prose? Extract one formatted for Midjourney, SD, and GPT Image.
8 free outfit descriptions a day without signing up — then 1 credit each once you log in.