7 ms·

Shapecatcher: Draw the Unicode character you want

87 points by nnnnni 12y ago

imakesnowflakes 12y ago
It works. Good work.
- nnnnni 12y ago
  It'd be nice to take credit, but it's not my page =-)
nnnnni 12y 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 12y 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 12y 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.
drodgers 12y 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).
- nnnnni 12y 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 12y ago
  sounds like you split up the stem and the bowl.
- yincrash 12y ago
  I just drew a question mark, and it showed up as the first result with a score of 0.899462
  - nnnnni 12y 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?
    - embolalia 12y ago
      I've been completely unable to get it to recognize an inverted interrobang (⸘) with no luck, though. Still, very very cool. EDIT: Tried one more time and got it as the 9th suggestion.
    - dthunt 12y ago
      Probably not training "matching", given that there is no place where you indicate what the correct thing was. It could be gathering data about how people draw shapes in general, but it's not immediately obvious to me how much that can help.
      - nacs 12y ago
        > given that there is no place where you indicate what the correct thing was. Actually there is a "Good" and "Bad" voting feature for each result.
benvds 12y ago
i drew a penis and got the "male sign" :-)
- rjtavares 12y ago
  I got a "Downwards harpoon with barb left beside downwards harpoon with barb right"
- brotoss 12y ago
  First thing I drew too lol
lgas 12y ago
This is great, it's like http://detexify.kirelabs.org/classify.html http://detexify.kirelabs.org/classify.html but for unicode instead of latex.
teddyh 12y ago
Nice. I wish the GNOME Character Map¹ could do this. ① https://wiki.gnome.org/Apps/Gucharmap https://wiki.gnome.org/Apps/Gucharmap
ajb 12y ago
There is also http://www.nciku.com/ http://www.nciku.com/ for chinese characters.
danieltillett 12y ago
I just did an 8 and got nothing close, but wow there are a lot of interesting Unicode characters.
Springtime 12y 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.
byuu 12y 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.
- nnnnni 12y ago
  What do you think of nciku.com?
  - byuu 12y 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_ 12y 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.
  - byuu 12y 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 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 12y 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...
Flenser 12y ago
Some other useful unicode websites: http://www.amp-what.com/ http://www.amp-what.com/ http://www.fileformat.info/info/unicode/char/search.htm http://www.fileformat.info/info/unicode/char/search.htm http://www.fileformat.info/info/unicode/block/index.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/ http://unicode.johnholtripley.co.uk/ -- mobile unicode support tables http://unicodinator.com/ http://unicodinator.com/ http://apps.timwhitlock.info/emoji/tables/unicode http://apps.timwhitlock.info/emoji/tables/unicode -- emoji :D http://character-code.com/ http://character-code.com/ http://panmental.de/symbols/info.htm http://panmental.de/symbols/info.htm and of course: http://copypastecharacter.com/ http://copypastecharacter.com/ Want more unicode resources? There's a list of other resources here: http://joewlarson.com/blog/2014/01/01/useful-unicode-resources/ http://joewlarson.com/blog/2014/01/01/useful-unicode-resourc...
- Flenser 12y 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 12y ago
  May as well plug my basic, bare-bones library too: http://charbase.com/ http://charbase.com/
rootbear 12y 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 12y 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 12y 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 12y ago
This is one of the first sites I learned about from HN a few years ago. Extremely useful.
adultSwim 12y 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 12y ago
  I got # on the first try. You helped the machine

87 points by nnnnni 12y ago

imakesnowflakes 12y ago

It works. Good work.

nnnnni 12y ago
It'd be nice to take credit, but it's not my page =-)

nnnnni 12y 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 12y 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 12y 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.

drodgers 12y 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).

nnnnni 12y 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 12y ago
sounds like you split up the stem and the bowl.
yincrash 12y ago
I just drew a question mark, and it showed up as the first result with a score of 0.899462
- nnnnni 12y 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?
  - embolalia 12y ago
    I've been completely unable to get it to recognize an inverted interrobang (⸘) with no luck, though. Still, very very cool. EDIT: Tried one more time and got it as the 9th suggestion.
  - dthunt 12y ago
    Probably not training "matching", given that there is no place where you indicate what the correct thing was. It could be gathering data about how people draw shapes in general, but it's not immediately obvious to me how much that can help.
    - nacs 12y ago
      > given that there is no place where you indicate what the correct thing was. Actually there is a "Good" and "Bad" voting feature for each result.

benvds 12y ago

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

rjtavares 12y ago
I got a "Downwards harpoon with barb left beside downwards harpoon with barb right"
brotoss 12y ago
First thing I drew too lol

lgas 12y ago

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

teddyh 12y ago

Nice. I wish the GNOME Character Map¹ could do this. ① https://wiki.gnome.org/Apps/Gucharmap https://wiki.gnome.org/Apps/Gucharmap

ajb 12y ago

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

danieltillett 12y ago

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

Springtime 12y 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.

byuu 12y 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.

nnnnni 12y ago
What do you think of nciku.com?
- byuu 12y 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_ 12y 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.
- byuu 12y 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 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 12y 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...

Flenser 12y ago

Some other useful unicode websites: http://www.amp-what.com/ http://www.amp-what.com/ http://www.fileformat.info/info/unicode/char/search.htm http://www.fileformat.info/info/unicode/char/search.htm http://www.fileformat.info/info/unicode/block/index.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/ http://unicode.johnholtripley.co.uk/ -- mobile unicode support tables http://unicodinator.com/ http://unicodinator.com/ http://apps.timwhitlock.info/emoji/tables/unicode http://apps.timwhitlock.info/emoji/tables/unicode -- emoji :D http://character-code.com/ http://character-code.com/ http://panmental.de/symbols/info.htm http://panmental.de/symbols/info.htm and of course: http://copypastecharacter.com/ http://copypastecharacter.com/ Want more unicode resources? There's a list of other resources here: http://joewlarson.com/blog/2014/01/01/useful-unicode-resources/ http://joewlarson.com/blog/2014/01/01/useful-unicode-resourc...

Flenser 12y 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 12y ago
May as well plug my basic, bare-bones library too: http://charbase.com/ http://charbase.com/

rootbear 12y 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 12y 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 12y 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 12y ago

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

adultSwim 12y 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 12y ago
I got # on the first try. You helped the machine