NetNews Usenet Archive 1993 #1/NN_1993_1.iso/spool/comp/ai/neuraln/4722

Internet Message Format | 1993-01-06 | 3.1 KB

Path: sparky!uunet!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!The-Star.honeywell.com!umn.edu!news.cs.indiana.edu!bsu-cs!bsu-ucs.uucp!01jmbrown
From: [e-mail]
Newsgroups: comp.ai.neural-nets
Subject: Re: Kahaner Report: Facial classification by neural nets
Message-ID: <1993Jan6.135012.13411@bsu-ucs>
Date: 6 Jan 93 18:50:12 GMT
References: <[e-mail]>
Followup-To: comp.ai.neural-nets
Organization: Ball State University, Muncie, In - Univ. Computing Svc's
Lines: 54

In article <[e-mail]>, [e-mail] (Henrik Klagges) writes:
> [Followups to comp.ai.neural-nets please.  -- Rick]
>
> [e-mail] (Rick Schlichting) writes:
>
>> Prof. S. Tamura at Osaka University Medical School, and Asst. Prof.  H.
>> Kawai ... have demonstrated that the back propagation neural network can
>> identify with high accuracy the sex of persons from their faces.
>
> That is a significant claim.
>
>> The test indicated that people have an identifying ability poorer than
>> the neural network [sic].  How the neural network differentiates between
>> the sexes has to be elucidated.  The neural network may extract facial
>> features that people cannot detect.
>
> The fact that a vanilla bp performs better than humans on a complex task
> at which humans normally excel makes it probable that the nn's success is
> based on 'artefact' information. Such artefacts could be easily intro-
> duced into the training set by e.g. taking the photos of the men & women
> in separate sessions with e.g. different background lighting. A very
> simple averaging calculation could then be performed by the nn to
> distinguish between 'male' (== background A) & 'female' (== background B).
>
>> ... it answered 93% correctly even for given
>> 8x8-array mosaic images, which were not recognizable as human faces to
>> people.
>
> That is a picture considerably smaller than a workstation icon, and the
> latter already require lots of clever pixel editing to resemble something
> like a mailbox (even if graylevels are allowed). I find it hard to believe
> that the sex of a face can be reliably inferred from a 8x8 picture _at all_.
> This performance claim makes an 'artefact'-based explanation very likely.
>
> Cheers, [e-mail]
>
>
> PS: The Kahaner reports are typically very useful and worth reading, I just
>     had to comment on this one.
> PPS:If bp really does the job, send me the code - they might give me a PhD
>     for it here :-) !


I've heard before that males have bones in their forehead that distinguish them
readily from females; males have a squarer shape (almost corners about where
the outside edges of the eyes are, whereas the female forehead is rounder.
Assuming this biological notion is true, I could see it _very_ possible to
distinguish males from females, *especially* if the subjects to be viewed were
placed in a somewhat darker room with the light limited to one bright source
from the side, a lighting technique which generally adds contrast to facial
features, making those features much easier for a computer image-scanner to
recognize.

-- Jeff Brown / [e-mail]