7. Attraction
Other people are the most important part of the world and our experience of it, so as these second sets of eyes watched our lives along with us, human forms would pop up often. Facial recognition is getting better all the time, and word on the street has it that the next version of iPhoto will include a feature which can locate, and tag people in photos automatically after manually telling it who’s who, allowing for easy searches: “This is Dominique- now find all pictures of her,” etc.
The use of this feature in real time is going to have a lot of applications- many of which will actually clue us in to the visual part of our subconscious more than would have ever been possible before this type of technology. First of all, people will want to have the computer keep track of who they find attractive in order to search for other attractive pictures and people:
“We become what we behold.”
Shallowness Profile (SP):
Basically an ever expanding record of what features you find physically attractive in other people. It's a permutation of the Netfix Model (NM) and works through biometrics and statistical analysis of common anatomical features and characteristics. Functionally, you could imagine it running in the background of any browser just like many applications already do. See below:
Shallow Practicality
3.2009
From essay on Digitalization:
Let's take a look at how this system would work: There would need to be a visual analyzer which would be able to identify and differentiate anatomical features. Think biometrics: it would understand things like facial symmetry, bone structure, skin clarity, baldness, disfigurements, hair color, skin color, muscle size, build, fat index, facial expression, and basic physical health. There's already been much research on this subject and the machines are improving.
So in theory just as netflix and pandora begin to understand your media tastes based on certain artificially imposed but nonetheless useful criteria (how the hell do you define jazz?), this application would attempt to understand your aesthetic taste in people- your idea of physical beauty. As you traveled through the internet, encountering pictures of your friends and complete strangers (advertisements, blogs, facebook), you would simply tell the application who you find physically un/attractive and perhaps your best guess at a reason why ("I hate his smile" etc.). Interestingly, after some time, through recording common physical features, this formula could "understand" things about your perception of beauty that you couldn't even tell it yourself. You might upload a picture of your mother, and label it as such. A year later the formula might realize that the characteristic bump on her nose is one of the features that most repels you in women. Or perhaps it will discover that you're attracted to people with similar eye movement, through the input of many pictures of the same people with different facial expressions which allows the computer to understand underlying muscle movement. This whole process could also be integrated with video. While functioning in real time it might find that your eyes (being watched by sensors synced to cameras on your glasses) are always drawn to men's hands. Realizing this, it might find similarities in the people you claim not to like. Although you assumed it was their personalities, it was in fact the way they all move in a jerky, neurotic way...
Understanding the way we judge people based on their looks wouldn't necessarily be a bad thing. It might clue us into the fact that we often harbor disdain for people based on illogical reasons. As a historical example, as soon as it became widely accepted that race had nothing to do with innate mental ability, we saw a decline in the amount of "scientific" treatises trying to defend such racist claims, which in turn gave the general populous less reasons with which to justify their racism. So realizing that I hate my roommate (partially) based on the trivial fact that he has my step-father's frown, I might be inclined to cut him a bit more slack. In cutting him more slack I might actually start to like him, which would establish in my mind a positive association for people with that type of face, thus allowing me to hate my step-father for what he is inside instead of simply for the way he contorts his lips. This would work in positive and negative ways accross the board having a whole range of consequences, but first and foremost it would allow us to better understand our own visual triggers. Personal attractions could be manipulated like never before.
Prologue -1 - 2 - 3 - 4 - 5 - 6 - 7 - 8 - 9 - 10 - 11 - 12 - Index and Short Summaries
http://research.microsoft.com/en-us/groups/vgv/We are interested in predicting a person’s facial attractiveness
in a given image. Generalized notions of beauty are subjective. However, an individual’s or group’s notion of beauty is often consistent and can be learnt. In the special case of faces, recent research suggests that there might even be a common, universal perception of beauty. Various factors, ranging from the evolutionary to the social and cognitive, have been attributed to explain the consistency in ratings between human subjects. Given training data in the form of photographs of faces along with their attractiveness ratings, our goal is to come up with features and a regression function which can help predict facial attractiveness in new images.
http://research.microsoft.com/en-us/groups/vc/
Photo Album Management – Face Tagging (Fang Wen, Jian Sun): Nowadays, more and more people take huge amount of photos in their daily life. The final goal of the photo album management work is help users to manage, search, share and make fun from these photos easily. ‘Who is in the photo’ is a good clue to organize and share photos. However, tagging people name is a tedious job for the user. Our Face Tagging work is trying to combine state-of-art face recognition and clustering technologies with a friendly user interface to make tagging effortless and fun.