Computational Perception [study realm]
+2
evil-mashimaro
sweet.
6 posters
Page 3 of 3 • 1, 2, 3
- JellyFish
- Messages : 108
Date d'inscription : 2007-10-25
Re: Computational Perception [study realm]
Mon Apr 24, 2017 12:58 am
There are some that you mentioned yourself.
- There are two eyes and two ears.
- Eyes have lens and pupil, ears have head, auditory cannal, eardrum...
- Eyes have photoreceptors that are sensitive to light, cochlea hair cells sensitive to sound.
- Both have ganglion cells, shaped like gabors.
- Eyes distinguish from left and right field of view, ears distinguish from lower and higher frequencies.
- Eyes have spatio-temporal receptive fields, ears have spectro-temporal receptive fields.
- Eyes have retinotopic map, ears have tonotopic maps.
- Dark
- Messages : 232
Date d'inscription : 2007-11-17
Re: Computational Perception [study realm]
Mon Apr 24, 2017 1:05 am
You can also have auditory and optical effect analogies like optical blur which is caused by the accommodation of the lens can be analogous to the cone of confusion caused by the HRIR.
Re: Computational Perception [study realm]
Mon Apr 24, 2017 1:09 am
Isn't it actually quite amazing that both eye and ear split the input data in half? WE ARE BIOLOGICALLY ENCODED TO PARTITION THINGS INTO TWO POLARITIES ~~~~
Re: Computational Perception [study realm]
Mon Apr 24, 2017 11:50 am
Is it actually true that sound is split in a way that high frequencies is used to detect level difference while low frequencies is used to detect timing difference?
Re: Computational Perception [study realm]
Mon Apr 24, 2017 11:52 am
No, it is all speculation~~~ the truth lies in the secret of sacred geometry ohohohoho
- JellyFish
- Messages : 108
Date d'inscription : 2007-10-25
Re: Computational Perception [study realm]
Mon Apr 24, 2017 12:12 pm
sweet. wrote:Is it actually true that sound is split in a way that high frequencies is used to detect level difference while low frequencies is used to detect timing difference?
It's a theory. High frequency has no information on timing difference because the wavelength is too small and peaks can shift in any direction to match in the coincidence detection structure. But we can still detect timing difference of high frequency for noise and impact sound because we can use the envelope of the sound.
Re: Computational Perception [study realm]
Mon Apr 24, 2017 12:13 pm
What about level difference? Why low frequency doesn't have information on that?
- JellyFish
- Messages : 108
Date d'inscription : 2007-10-25
Re: Computational Perception [study realm]
Mon Apr 24, 2017 12:14 pm
Eh... the slides and exercises did not talk about it. :/
Re: Computational Perception [study realm]
Mon Apr 24, 2017 12:17 pm
It's because people speculate that level diff is log scaled! IT'S THE POWER OF THE LOG THAT SPLITS THINGS INTO TWO~~~~ see we also use log in statistical classification. WHAT A SNEAKY FUNCTION
Re: Computational Perception [study realm]
Mon Apr 24, 2017 12:20 pm
Oh, so at higher frequency, the ratio is smaller and the log value would have bigger changes, whereas at lower frequency, the ratio is larger and the log value would have little change. Thanks, evil-mashimaro <3 Let's study well together
Re: Computational Perception [study realm]
Mon Apr 24, 2017 12:23 pm
No problem my sweet. But more precisely, you mean the slop of the log rather than the change of log. If all grad students write like you, I'm sure someone will kill himself. ohohohohoo
- JellyFish
- Messages : 108
Date d'inscription : 2007-10-25
Re: Computational Perception [study realm]
Mon Apr 24, 2017 2:49 pm
Seems like that someone is not coming back.
- Spirit
- Messages : 163
Date d'inscription : 2007-11-17
Re: Computational Perception [study realm]
Mon Apr 24, 2017 4:15 pm
dat dude wont cuz he ll c m wright like sht ag'in
- shadow
- Messages : 261
Date d'inscription : 2007-10-25
Re: Computational Perception [study realm]
Mon Apr 24, 2017 4:35 pm
And you thought he would leave you talk like sht?
Re: Computational Perception [study realm]
Mon Apr 24, 2017 4:37 pm
SHADOW~~~~ wait, you need to enter from the front door too...
- Spirit
- Messages : 163
Date d'inscription : 2007-11-17
Re: Computational Perception [study realm]
Mon Apr 24, 2017 4:41 pm
so he dared to show up...
Re: Computational Perception [study realm]
Mon Apr 24, 2017 4:43 pm
Spirit, did you do anything to make him come?
- Spirit
- Messages : 163
Date d'inscription : 2007-11-17
Re: Computational Perception [study realm]
Mon Apr 24, 2017 4:44 pm
He just came...
...after reading my post.
...after reading my post.
- Spirit
- Messages : 163
Date d'inscription : 2007-11-17
Re: Computational Perception [study realm]
Mon Apr 24, 2017 4:49 pm
datz haw mush he mis mi
Re: Computational Perception [study realm]
Mon Apr 24, 2017 9:22 pm
TOMORROW IS THE END OHOHOHOHOOH
GOOD LUCK EVERYONE
GOOD LUCK EVERYONE
Page 3 of 3 • 1, 2, 3
Permissions in this forum:
You cannot reply to topics in this forum