The Hexacoto

Listening to the sound of one hand clapping

Category: Art

You ain’t so good, Google Neural Net

Google recently launched an interactive web game to train its neural network to recognise objects. The game, Quick Draw, calls on human users to draw a prompted object within a short period of time and the machine tries to guess what it is based on what it has learned so far from all of the inputs of previous players. Quite ingenious, to crowdsource training a machine learning (ML) program since many people are always looking for an excuse not to do work.

I wanted to test its learning limits. I more or less had a sense of how previous inputs for the prompts would look like, since humans tend to draw objects similarly when under time pressure. I wondered if I drew all of the objects from a different perspective, would the program still recognise it as the object — a task which humans are very capable of?

The answer is: not really.

I experimented with drawing in a sequence that would not be obvious what the object is immediately, but the end product would be discernibly apparent. I experimented with odd and skewed perspectives. Google Neural Net failed most of the time.

screen-shot-2016-11-17-at-3-17-20-pm

I think my mountains were really good! I started with a skewed line to not trigger immediate “mountain” responses from Neural Net and then quickly added half lines but by the time I was done, any human would have seen that these are really good mountains.screen-shot-2016-11-17-at-3-30-52-pm

I thought it was cute that my mouth were interpreted as a bear, an owl and a smiley face. What?? I started with the top line forming the nose, lips, mouth, chin and neck, followed by the back of the head. I filled in details and drew an arrow pointing to the mouth. In Neural Net’s fairness, its creators probably never accounted for it to learn the concept of pointing — a task that I don’t think is too difficult given how far we’ve come along in ML. It seems Neural Net has really only been learning to identify objects by scanning them as a whole.screen-shot-2016-11-17-at-5-23-15-pm

I drew a jagged tooth key, instead of a wedge-end key because I thought it’d be too obvious. By the time I finished the key, Neural Net still hadn’t recognized it. I had some time left and literally drew in the words “KEY” hoping it’d help Neural Net along but noooope. It thought it to be a crocodile. Cute croc though.

Looking at what examples Neural Net uses as its learned base to pass judgment, one sees that humans tend to draw things either profile or head-on, and hence how Neural Net learns to identify objects.

screen-shot-2016-11-17-at-5-21-49-pm screen-shot-2016-11-17-at-5-22-04-pm

Come on. My butterfly was clearly the best butterfly all of Neural Net’s learned examples.screen-shot-2016-11-17-at-5-22-35-pm screen-shot-2016-11-17-at-5-22-45-pm

How are some of your examples even mushrooms!? They look more like penises! I declare my mushroom to be mushroomier than your learned examples!

Google Neural Net, it seems you have a long way to go.

 

Bonus pic from a friend:

15129874_10154732065471913_1577713233_n

COME ON. In what universe are these trombones?? I’m starting to think people have never seen what a trombone looks like.

Balance Your Chicken

Sometimes, you just gotta lighten up and take a break and create pointless, stupid stuff. I had been on the train, on my phone, extracting bridges from songs for future use on short video clips like these when suddenly the caption “Fried chicken can be part of a balanced diet too!” popped into my head. And as a fan of Cibo Matto, the idea of for this video clip came to fruition.

And I realise I’m no stranger to creating stupid things to make sense of life.

I probably made this back in 2008 or 2009. There were a bunch of other clips from that same filming session that I am not releasing because they’re very cringeworthy.

Reddit Kim

pisai

So I made something, after having been inspired a reddit comment that said that Singaporeans would never waste money buying reddit gold for others. Hence I came up with this idea, for those kiamsiap (stingy) Singaporeans when they don’t want to buy the real thing. (Well it’s not like reddit gold is of any use anyway)

Other ideas included the nose picker projectile flicking the booger, culminating in a giant explosion. But who has time to sit down and draw/animate all that?

Made with 25 stills drawn by mouse using Adobe Photoshop, animated with Premier Pro. Done in one sitting from 8pm to 3am.

Listening Comprehension: Fill in the empty spaces. (27 marks)

Lines

I cast a line, and I waited, not expecting a response.
But, lo! A bite!
You called me by a familiar name, a name from a time thought foregone.
I reeled; could it be?
That no matter the time lapsed, distance spanned, silences met,
that things were as they should have been?
But I knew better: time ravages, distance cools, silence forgets
And pulled in to meet with trembling hands and a clamoring heart.

Captain

You spoke of how you thrashed, how you were lashed
by forces that had you trundled, trussed, and tried.
Blinded beyond belief, beyond benefaction — barking madness.
Bye,” you said, as I remembered, back then,
back when you cast yourself afloat.
And I cried as I heard your tale,
sorrowed that I could not sail with you
during your stormy seas.

Colours

“Have you heard about colours?” you asked.
“They can be quite therapeutic,” you said
through a new veil of age and wisdom.
And so I sat in silence for hours,
drawing out lines and figures
dashing out spaces
despite my cramping hand
(whilst you worked),
so that I could give you something to fill in with colours
to repair the missing gaps in our lives.

20150909_145534

The TRUE reason why this blizzard is scary

stonest

#noescape

We danced, we laughed, we skipped around. And then we were twenty-six.

 

 

I wish you good luck.

I wish you good luck.

Captain

We became friends, and we were friends, and then one day, friends no more.

What does one do when a friendship, whose tethers are time-worn and frayed, comes loose and slips away? Time ebbs, and the vessel departs, do I fling myself to reel it back?

Or do I set it alight and let it go, in a Viking’s funeral, remembering that it once burned, with the last memory of its light in sight?

And now that I find myself aware of all the ropes around me in varying stages of decay, do I darn them, mend them, let them be?

Lines

“Let it go,” it says. “Let it go. It is the way of life. Two parallel lines may never meet, but if they differ enough, will remain close enough that they merge for a really long time. But eventually, they will depart, and then it is time to go.”

“But why does it have to be that way?” I ask. “Our lives are not simple straight lines. We meet by circumstance, but it is by virtue of entanglement that we remain hurtling through space bound; entwined.”

“All things tend towards chaos,” it says. “And in chaos squared, tangled lines come unwound, and come free of each other. That is the very essence of life. A static line is a dead line.”

Inexorable

We must grow up, but must we grow apart? Perhaps part of growing up is learning to let go, perhaps part of letting go is to know — when to say hello; when to say good morrow; when to say good bye, and say no more.

Perhaps one day, I will hear from you again. Until then, fare thee well, I’ll keep these memories.

shaf

Ceci n’est pas un concombre

 

HikaruCho6

HikaruCho7

Japanese visual artist Hikaru Cho is known for her transformative paintings — turning things into something else through her application of paint. In this series, she turns various food items into looking like other food items, such as the banana above into a cucumber. She also did the same with turning an egg into an eggplant, and a tomato into an orange.

But Hikaru Cho does really amazing (and creepy) work with transformative painting on humans. Here are some of her best works:

Her site’s definitely worth checking out!