Shapecatcher: Draw the Unicode character you want

by nnnnni • 12 years ago

87 38 12 years ago

nnnnni5• 12 years ago

It doesn't support every unicode character yet, but it's getting there. For example, it recognized the Kannada character ttha (ಠ), but it doesn't know the poop (💩) character.

rubinelli• 12 years ago

It does! I got it as one of the suggestions while trying to make it recognize the snowman (my go-to character when I stumble on one of these tools).

ath0• 12 years ago

Did you ever get it to recognize the snowman? That's my go-to also. I've tried two snowballs, three snowballs, with and without arms, with and without stars, with and without hat... and all I keep getting is tongue. I am pretty bad at drawing though.

benvds• 12 years ago

i drew a penis and got the "male sign" :-)

rjtavares• 12 years ago

I got a "Downwards harpoon with barb left beside downwards harpoon with barb right"

brotoss• 12 years ago

First thing I drew too lol

lgas• 12 years ago

This is great, it's like http://detexify.kirelabs.org/classify.html but for unicode instead of latex.

teddyh• 12 years ago

Nice. I wish the GNOME Character Map¹ could do this.

① https://wiki.gnome.org/Apps/Gucharmap

ajb• 12 years ago

There is also http://www.nciku.com/ for chinese characters.

danieltillett• 12 years ago

I just did an 8 and got nothing close, but wow there are a lot of interesting Unicode characters.

Springtime• 12 years ago

What would perhaps be a worthwhile feature is the ability to input already known unicode names into a text box after seeing the results, to feed the database with more useful matches.

For example, tried drawing a somewhat joined 'TM' a couple times but no matches for 'Trademark symbol', however a way to manually input that unicode/name might provide the database with a positive match for the next user trying to find it.

imakesnowflakes• 12 years ago

It works. Good work.

nnnnni5• 12 years ago

It'd be nice to take credit, but it's not my page =-)

drodgers• 12 years ago

They didn't think to weight the prior probabilities by usage frequency* - drawing a reasonable ? gives me ȓ, ᕉ, ╔, ᣑ, Ѓ, ק, ᒌ, ŕ, ᒤ, ᒦ, ņ, ᒯ, ѓ, and finally ?.

I'm also guessing that they're directly comparing the handwritten character to some version of the unicode character rather than with human attempts to draw the character. Human drawings are often quite different (more slanted, stylised etc.) than typeface characters. This is much more forgiveable though because assembling a good dataset for human drawn characters is hard (especially for any reasonable chunk of the unicode set).

(*this is fairly easy to do: just find some large source of typical unicode, like Wikipedia in all languages, and index them).

nnnnni5• 12 years ago

Yeah, it seems like they're taking the drawing and matching the edges and points up with edges and points in a font file. Hopefully they'll start weighing by frequency and add more glyphs soon!

Hello71• 12 years ago

sounds like you split up the stem and the bowl.

yincrash• 12 years ago

I just drew a question mark, and it showed up as the first result with a score of 0.899462

nnnnni5• 12 years ago

Yeah, I just tried it with the inverted question mark (¿) and interrobang (‽) -- both attempts had the correct glyph for the first result.

Maybe we're training it?

GregBuchholz• 12 years ago

They didn't think to weight the prior probabilities by usage frequency

I don't know if that is the right metric for this sort of tool. I'd guess the use case is for trying to find infrequently encountered characters. It should probably try to detect your current locale, and then say eliminate all ASCII characters when you are suspected of speaking English, etc., since you are already aware of how to type a question mark.

byuu3• 12 years ago

Oh wow, the drawing mechanism is really satisfying.

Too bad about not supporting 漢字. The only half-decent IME pad is on Windows. Online ones (kanji.sljfaq.org) and Xorg ones (ibus-mozc) are just horrendously bad at detection. I usually have to resort to multi-radical lookups.

nnnnni5• 12 years ago

What do you think of nciku.com?

byuu3• 12 years ago

(Late response, sorry.) Seems focused on Chinese, but Han unification means it's still somewhat useful. The drawing feels clunky and I see both my mouse cursor and their custom pen cursor at the same time. The drawing window is rather small. Prediction seems worse than Windows IME, but better than mozc IME.

ue_• 12 years ago

Do you mean multi-part? Each kanji only has one radical, but composed of one or more "parts". jiso.org has this lookup feature.

byuu3• 12 years ago

(Late response, sorry.) It may not be proper to say this, but if you look at 加, to me it's clearly the 力 radical + the 口 radical. I know that officially it's classified under 力. wwwjdic has a tool based around my observation, called multi-radical lookup: http://www.edrdg.org/cgi-bin/wwwjdic/wwwjdic?1R

Sometimes it won't include one of the radicals that are clearly there, though. But this tool is frequently helpful to me when IME pads fail me, and stroke counts are fuzzy.

xg15• 12 years ago

An interesting side effect of this is that it shows once again why (naive implementations of) international domain names presented such a large security risk. Just draw an "A" and look at the results...

Flenser2• 12 years ago

Some other useful unicode websites:

http://www.amp-what.com/

http://www.fileformat.info/info/unicode/char/search.htm

http://www.fileformat.info/info/unicode/block/index.htm

(I've added the above as search engines in Chrome with short mnemonics for the keyword.)

There's also:

http://unicode.johnholtripley.co.uk/ -- mobile unicode support tables

http://unicodinator.com/

http://apps.timwhitlock.info/emoji/tables/unicode -- emoji :D

http://character-code.com/

http://panmental.de/symbols/info.htm

and of course:

http://copypastecharacter.com/

Want more unicode resources? There's a list of other resources here:

http://joewlarson.com/blog/2014/01/01/useful-unicode-resourc...

Flenser2• 12 years ago

Forgot to mention, I use this bookmarklet with amp-what to change the font to Segoe UI Symbol as that often has more Unicode symbols and is the font used in a lot of the places I tend to use unicode symbols.

      javascript:(function(){var newcss="samp { font-family: Segoe UI Symbol; }";if("\v"=="v"){document.createStyleSheet().cssText=newcss}else{var tag=document.createElement("style");tag.type="text/css";document.getElementsByTagName("head")[0].appendChild(tag);tag[(typeof document.body.style.WebkitAppearance=="string")?"innerText":"innerHTML"]=newcss}})();

mmastrac• 12 years ago

May as well plug my basic, bare-bones library too: http://charbase.com/

rootbear• 12 years ago

I've often thought that the best way to get access to the richness of Unicode would be a drawing pad, perhaps as part of the keyboard, or as an on-screen area, for use with a mouse. Character maps just seem clumsy to me.

For fun, I tried Eth ð, Thorn þ, and Hungarian ű, all of which it got, but not as the first choice. It did not find the Ing rune, which looks a bit like a < and > combined.

whitten• 12 years ago

This is pretty cool. I expect that even if the code is open source, that the real value is in the dataset used. Does anyone know the licensing information ?

lost_name• 12 years ago

It seems (and is logical, I suppose) that you have to match the symbol pretty closely to get what you're looking for. I drew a car twice, the first time getting absolutely nothing relevant, and the second time -- trying to be more precise and using all the space available -- got automobile, taxi, bus, etc, etc.

edit: The primary problem here, I mean, is that if you don't know what the symbol looks like and want to see if it exists, you might not get hits the first time you try to draw, but it might not actually exist anyway.

danielweber• 12 years ago

This is one of the first sites I learned about from HN a few years ago. Extremely useful.

adultSwim• 12 years ago

I drew #. Returned a bunch of characters but none of them was regular old 0x0023.

Took me several tries to get # (regular ascii number sign - i.e. shift+3).

kitwalker12• 12 years ago

I got # on the first try. You helped the machine

Shapecatcher: Draw the Unicode character you want

Related Stories