Personally, I'm all for the use of amusing (arousing?) test images, as long as they exhibit features that actually make for a good test image (a mix of flat and textured regions, sharp boundaries, etc.).
The original Lenna image is still copyrighted and will remain so for a while (even if right now Playboy Enterprises, Inc. has decided to not sue people over its use in research) and people can't seem to agree on a standard replacement [1].
Since that is the case, how about a Kickstarter campaign to recreate the 512x512 image as close a possible and distribute the imitation under a free license? I.e., you could hire a professional model (by running a Lenna double contest online?) and a pro photographer, recreate the environment and the props, etc. Keep it clean (no remaking the full nude) and you could get free press and support from the researchers themselves.
I wish he'd post the scan of the original, rather than the JPG of the scan of the original. Then we might be able to use it for compression research.
It's not about wanting to see nudity. The original is useful because humans are hardwired to recognize nudity. It's not a coincidence that tasteful nude pictures are considered fine art. There are literally a hundred thousand tiny features that are lost by cropping the image to her face and shoulders. Those features are much easier to compare than, say, pictures of nature. Artifacts introduced into pictures of leaves and grass are far less noticeable than artifacts introduced in the subtle variation of the colors across her back.
Surely playboy keeps the negatives for all their materials. Might be possible to reach out to them for a high quality scan.
Anyways, you're absolutely right, people's brains are very very very much optimized for evaluating the microfeatures in a person's appearance, so its actually the right perceptual way to benchmark lossy compression (or at least one very good way).
I don't understand your comment. Why would this nude, of all of the nude images that exist, have any special importance in compression research? Other than historical reasons, of course.
For example, the colors and registration will be different than the original image, so it's not like a new scan will identically include the same pixels as the old scan. If a new scan is from the paper copy, then there might have been a change in the ink colors over time. If the new scan is from the negative then there's a difference between that color and what was printed.
You claim that "humans are hardwired to recognize nudity." Is that really true or is it a hypthosis? How was this tested? What is the signal difference between fully nude and partially nude? In the Lenna case, she is not completely nude - how will that confound the results? Does wearing a bikini affect the results? What about gloves? A cummerbund?
I grew up in a white-dominated culture and have more experience with white skin. I've noticed that it's easier for me to identify white skin than black skin. This suggests that recognition has a strong learned component. In those tests, how was the influence of that learned behavior disentangled?
Is the 'hardwired' recognition due specifically to nudity or is it to the transgression of cultural norms? That is, some cultures de-emphasize nudity as being something other than just a person without clothes. This is not generally true of US culture. I can conjecture that there is a hard-wired system to see if a person is following cultural norms, so in the US, nudity would stick out more than in cultures where casual nudity is more common.
Now that I think about it, any hardwired signal like this should be easy to test. Have an array of, say, 10x10 photographs, where one is nude and the others are wearing appropriately skin-tone clothes (fully dressed, partially dressed, etc). Then time how long it takes to find the nude person. This test can be done across multiple cultures pretty easily, so long as nude images are tolerated.
This is similar to the studies I've read about done on facial expressions, to show that some are very cross-cultural and thus likely more hard-wired, while others are not and thus more likely to be learned. For example, as your baseline reference, use a similar 10x10 grid but where one person is frowning while the others have a neutral face, and time how long it takes to find the frowning person.
Since it sounds quite simple to carry out, I figure it's been done, so if that's the sort of work you're referring to, can you kindly point it out? A Google search gets bogged down with references to algorithm for detecting nudity in images.
Sadly not the uncropped version. There's a high-res version of the uncropped original, but it's JPG, which is inherently unsuitable for compression testing.
They go on to say that the photo had to be cropped due to the equipment they were using.
"The Muirhead had a fixed resolution of 100 lines per inch and the engineers wanted a 512 x 512 image, so they limited the scan to the top 5.12 inches of the picture, effectively cropping it at the subject's shoulder."
a 512x512 pixel scan of a printed image really isn't ideal source material. Nowadays you definitely want something larger, and to get that in good quality you need the negatives.
In the 90s, Lena was pretty much one of the images you spent a lot of time looking at if you were in compression research. It was part of a standard set of images we were always comparing agains.
I want to say the last time I ran across mention of Lena was either here on HN or through a rant of Charles Bloom -- http://cbloomrants.blogspot.com
I do find the story fascinating each time I read it.
"Two reason was stated in "A Note on Lena" by David C. Munson. First, the Lenna image contains a nice mixture of detail, flat regions, shading, and texture that do a good job of testing various image processing algorithms. It is a good test image! Second, the Lena image is a picture of an attractive woman. It is not surprising that the (mostly male) image processing research community gravitated toward an image that they found attractive. "
It's perfectly reasonable that image processing and television are optimized for accurate skin tones. But they are optimizing for light skin tones by using test images such as Lenna, the BBC's test image [A], and the BBC's "Flesh Tone Reference Card" [C]. I think this harms image quality for dark skin tones.
Television broadcast systems include automatic skin tone correction that "looks for colors in a specific [colorspace] area and any colors within that area are made a color that is closer to the skin tone." There's actually a "correct" skin tone and colors are adjusted to be closer to it. This is implemented in chips [D] for TV signal processing and even at the LCD panel controller level.
I've noticed reproduction of dark skin colors is much worse on video conferencing than light skin colors, so this isn't just a theoretical complaint.
Its not so much that the image is pornographic, the interesting part is that most computer graphics group is male. If there was a 50/50 male to female split, I doubt this image would have become the de facto test case.
Eh, as a graphics researcher, I'd say it's not a gender issue. Anyone who spends a significant amount of time researching the human visual system in a fair an unbiased way is inexorably pulled to the conclusion that comparing pictures of nude bodies is one of the major tests of whether you're achieving acceptable results, whether you're making a painting or developing a compression algorithm. Humans are hardwired to notice details in naked bodies, possibly moreso than faces (or at least we notice a different set of details than faces). They were onto something important when they chose the Lenna picture, and it's a shame that the original had to be cropped due to technical limitations. And it shouldn't be taboo to use a male model any more than a female one -- both are equally important.
Why are we talking about naked bodies when the crop of the Lenna picture that's actually used only contains a face? And a bare shoulder, but I don't think that's been pornographic for the last 100 years.
Do you want to have more female co-workers or not? Working with nude images when you don't really need to is a great way to discourage people from working with you.
Why? Does it discourage models from working with photographers or artists? There seems to be plenty of female photographers or artists that have no problem with nudity, too.
Your comment is hard for me to interpret. I would normally take it as a given that the culture and male/female dynamic in computer science research is (or at least ought to be) different than the culture in the modelling industry (and especially the nude modelling industry).
Ok, let me try to explain a bit better. The parent poster seems to say that women are somehow "put off" by nudity. That's a perfectly valid hypothesis.
But then art (and I am talking primarily about art, not modelling) is a problem, which requires further explanation. Because artists seem to be, at least to me, totally obsessed with nudity (and I think sex, too, but feel free to disagree here). So why that attracts women, in droves in fact, if they are put off by it?
My answer is that the hypothesis is wrong. Females are not put off by computer scientists, that's not the reason why they don't do computer science. In particular, I doubt you would find CS people more sexist than other group of men, I actually believe they are mostly nicer (and also lot more honest about their views, which can strongly bias it).
The reason why women don't do CS is they lack confidence and positive role models to do it. You can actually see the differentiation happening in high school, where no one really knows how it works on the workplace yet. And that's where, or maybe even earlier, the problem needs to be addressed. It's probably cultural and frankly, whether the cause is wrong beliefs of men or wrong beliefs of women is irrelevant for the solution (wrong here in the sense "differing from reality").
Edit: You could argue (I think you do) that the cultural norms about nudity in art and computing should be different. I don't disagree (in fact, they are, which makes it a moot point anyway). But then, isn't this a case of someone projecting their own cultural norms to somebody else? Is that really a failure of some culture that it doesn't respect your (rather arbitrary and harmless, compared to art) norms?
I don't know why things are different in the art world and it's a good question. But this sort of thing has been A/B tested for computer science [1]. Of course it takes more than one study to get scientific consensus but I'll take it over armchair reasoning.
The flaw in that study seems to me that for starters, boys don't need to be convinced to go into CS. And those stereotypes are mostly just stereotypes - there were no such things at my university, for example. It was a rather ugly building, though. But it' not that men prefer ugly buildings (I definitely didn't), but that they keen enough on those subjects to go for it anyway.
However if the finding is that if you make it nicer, more women will choose it, that seems rather trivial. Of course if you make something more attractive, more people will choose it.
Sure, it's just one part of the overall environment. If you wanted to decorate a room to say "boys club - girls keep out," the first thing you'd do is hang up some pinup posters. But in the fashion industry they work with lots of pictures of beautiful women and women outnumber men. And this is probably US specific too; I imagine they do things differently in France.
In computer science (either academia or industry) we have a recruiting problem and we're facing an uphill battle. We have to care about how it looks.
I don't think that the parent poster is saying "women are put off by nudity". (As you point out, there are other contexts where women can often be just as comfortable with it as men are.) As I read it, the parent poster is saying that women are less interested in working in professional environments where their colleagues expose them to nudity unnecessarily. (That word "unnecessarily" may be part of why things are different in art than in graphics research.)
Someone else has already shared some formal research on how other "in-group" signals have exactly this effect. I don't think it's much of a stretch to suggest that "public consumption of heterosexual-male-targeted erotica" could have a similar effect. But even without that research, the fact is that plenty of women have been very explicit that nude or sexual images in the workplace make them feel unwelcome, unequal, and in some cases even threatened. Regardless of the deep causes of those reactions, the experience is undeniably real for the people who report it. And that comes back to the parent poster's point: if your company environment welcomes nude or erotic imagery without good reason, many women will choose not to work with you.
Finally, you ask whether it's unfair for one culture ("women", broadly speaking) to project its own cultural norms onto another ("computing", in this case). I see two answers. First, the pure capitalism that some folks here are fond of: if your company's atmosphere tends to push away talented female employees and your competitor's does not, they'll be more successful at drawing on that talent pool, eventually giving them a big advantage. (One could probably extend that reasoning to entire industries or societies with a little work.) And second, a social justice perspective: not all norms are morally equivalent. The classic example is that slavery used to be a perfectly accepted norm, but few today would argue that it was wrong for the abolitionists to seek to change it. I'd like to think that one day our social context will have shifted enough that nude images in the workplace will have the same impact on men and on women, but we aren't there now and it's hard to imagine it happening in my lifetime (or even two or three generations down).
[Oh, and one more point, since it's something that took me years to catch on to: as a guy, I'm very cautious these days when it comes to deep theorizing about why women (or other underrepresented groups) might feel the way they do about issues like this. The thing is, while trying to reason out all this stuff is just an intriguing intellectual exercise for me, for women it's their constant reality. And let me tell you, there's no upside to taking someone who's obviously unhappy about something and trying to dispassionately pick apart what exactly about it is bothering them. In the end, it's almost always more productive to simply say, "Wow, thanks for letting me know you feel that way. What can we do to make it better?"]
My point was that the nudity in art is, for the most part, unnecessary. Yet people, including women, bear with it.
The point is, there is probably no absolute, specific reason (such as nudity) why women don't want to go to CS. The reason may well be the cultural difference from the norm (of majority of population) itself, no matter what it is. Whether it's nudity or Star Trek, doesn't matter. So the morality doesn't even come into play here.
And that's why positive examples are needed. To show, "OK, it's a bit different culture, but that's fine" sort of thing. (Also to show that girls can be good in CS, because they often seem to lack confidence in this - at least I met several that didn't feel they could be programmers, even though they were lot smarter than me.)
It wouldn't be controversial to say that humans are more prone to see green than blue or red, or more interested in faces than in leaves, so I hope graphics researchers won't be offended by the truth that preserving details in naked bodies is a wholly necessary benchmark for a compression algorithm, just like it is for a painting.
Okay, let me put it differently: I'm not an image compression researcher, but I find it hard to believe that you actually need pictures of naked women to judge compression quality. Pictures of people, sure, and maybe even showing some skin. But beach photos would work fine for that, as did the cropped picture of Lena back in the day. So: citation needed.
(If you were just aiming for humor here, my apologies for missing the point. I blame the usual issues of unemotional text.)
Whether or not any given woman is fond of looking at pictures of naked women isn't the point. The issue is that it's difficult to give women assurances that they are being treated equally as colleagues if the professional environment also embraces cultural norms that treat women as eye candy or playthings.
And this isn't an assumption, in any case. Plenty of women have been quite explicit about feeling uncomfortable about erotically charged images in the workplace. (You know this.) I'm sure some women think it's fine, too, but that doesn't make the feelings of the others any less valid.
"if the professional environment also embraces cultural norms that treat women as eye candy or playthings."
I think sexual attraction will always exist between the sexes, so any professional environment will have to deal with that. I don't agree that erotica imply that women are just playthings.
The image used was just the head of a model, btw.
Probably some men disagree about the use of the picture, too. Everybody is free to use another picture, though.
I wonder if some women feel uncomfortable about the display of an unveiled female head in a photograph - only veiled photographs would be really OK?
Lastly, what if I don't want to work with somebody who is uptight about that sort of thing? I think it should be possible to build a team of people who get along. Presumably there could be some workplaces with uptight rules, and some more lose workplaces. Why should "uptight" always win?
Look, it's not about the Lenna image in particular (even though it was the inspiration for this thread): given the cropping and the long history, I think there are much bigger issues for the community to worry about.
But it sounds like you're addressing the broader issues just like I was: you suggest that you might not want to work with someone who was "uptight about that sort of thing". All I can offer is this: I've heard multiple stories from female engineers at tech conferences who have had trouble getting attendees to talk to them because everyone assumed they were just there as some sort of booth babe. It can't be good for their companies to have an engineer's time wasted in that way, and it's certainly not good for the women's careers if they're unable to network as effectively as their male peers. I won't try to claim that I can nail down a specific causal chain, but it seems pretty clear that a culture that routinely uses women purely as attention-getting devices or for visual enjoyment would naturally tend to undermine women who work in technical roles in this way. And when you drive away or devalue half of the potential talent pool just because a few guys don't see it as a problem, that's a bad thing.
[Oh, and a side note to whoever is downvoting out there: https://news.ycombinator.com/item?id=3782124 I'm open to the suggestion that my comments in this thread have "provided no substance or insight", but I'd love to know why. If you just happen to disagree with my point (or anyone else's), that's not what the Hacker News downvote button is for.]
Use of booth babes is a somewhat different case. I personally don't like that practice, but it seems likely to me that women on tech conferences are assumed to be booth babes because it is so often true (apparently - I haven't studied that). That is, the likelihood of a woman to be at a tech conference an not be a booth babe might be low, so that the expectation is otherwise. That sucks, of course, but the only way I can see to change that is more tech women attending conferences. Or maybe some obvious signs of non-booth-babeness.
Also, did those women really not manage to talk to anyone? That seems odd, given that booth babes are probably also supposed to make people talk to them, and are considered an effective means for that?
> And it shouldn't be taboo to use a male model any more than a female one -- both are equally important.
Correct, and if there would be a tradition of using both, that would already be different. I see what you’re getting at, judging skin tones etcetera is important, but the Lenna picture doesn’t represent only nudity—she’s a playmate, and represents all the cultural luggage that comes along with that. If you’re looking for an inviting workplace open to others than cis straight males, making a playmate a central part of your research culture is silly.
Perhaps, but that wouldn't be because women don't like looking at pictures of naked women.
[G]ay men overwhelmingly become sexually aroused by
images of men and heterosexual men by images of women.
In other words, men’s sexual arousal patterns seem obvious.
...
In contrast to men, both heterosexual and lesbian women
tend to become sexually aroused by both male and female
erotica, and, thus, have a bisexual arousal pattern.
I am actually fairly sure the (cropped) image that is actually used appeals to women as much as to men. It could be straight from the cover of a fashion magazine.
Researchers decided to use a picture from a playboy, a magazine that thinks chicks nude are kinda all they are good for (come off it if you seriously think they don't)and we all think it's ok and want to continue it in the comments.
Yes, it's history, like rape and pillaging. Yes things of old are cool cause they happened ages ago. But get over it.
Nothing against Lenna, cause she was in the industry, but the fact anyone thinks it's cool to bring a playboy into a place of research offends me to the point of boredom.
So there is at least one scientific paper (image processing) out there which references Playboy (I wonder if Google Scholar has a citation index for this?)
http://arxiv.org/abs/1202.6429
(Additional discussion at http://www.kitware.com/blog/home/post/246)
Personally, I'm all for the use of amusing (arousing?) test images, as long as they exhibit features that actually make for a good test image (a mix of flat and textured regions, sharp boundaries, etc.).