Question
Asked 13 June 2018

Anyone out there knows the truth about facial similarity? I mean please! It must be more than a simple pixel-wise metric, right?

For many studies it is important to know how "similar" your images are. Simple pixel metrics surely dont reflect our perception. Is there any metric and method what is accepted to tell if the face of Scarlett Johanson e.g. is more similar to the face of Marylin Monroe or Ingrid Bergman???
Many thanks for any suggestions (with references).
Gyula

Most recent answer

Justin Gaetano
University of New England
Hello,
I may have some alternative answers - feel free to send me a private message.
Cheers, Justin

All Answers (4)

I can share a way with you, but it doesn't involve the type of research you already do. It's much different (and easier), it's also a bit of a trade secret tho, but if you want to know, pm me.
1 Recommendation
Ruben Garcia-Zurdo
Complutense University of Madrid
Hi, check this:
On Constructing Facial Similarity Maps
Best!
Justin Gaetano
University of New England
Hello,
I may have some alternative answers - feel free to send me a private message.
Cheers, Justin

Similar questions and discussions

Is there more to long-term memory than just memory?
Question
3 answers
  • Edward J TehovnikEdward J Tehovnik
Eric Kandel (2006) has revealed that the consolidation of memory at the level of the nucleus is a bipolar process: chemical agents exist in our cells that can either potentiate or suppress memory. Having a double-ended system prevents extremes: remembering everything or remembering nothing. As a child we are rewarded for remembering everything we are taught in school under the assumption that all knowledge is good. But what happens if the knowledge is tainted such as that the black slaves on plantations enjoyed being supported by the white slave owners, that the Holocaust was a fabrication, that the recent election in the United States was rigged, that vaccines produce massive side-effects, that drinking Clorox is an effective way to kill Covid-19, and so on. It is instructive that Albert Einstein was not a great student (i.e., did not like to memorize things and he had difficulty in his second language, French, which he needed to complete his university entrance exams, Strauss 2016) yet his ability to zero-in on the important data while excluding nonsense is what made him an extraordinary scientist. Ergo, the management of one’s memory may be as important as having a good memory.
References
Kandel ER (2006) In Search of Memory. The Emergence of a New Science of Mind. W.W. Norton & Company Inc., New York.
Strauss V (2016) Was Albert Einstein really a bad student who failed math? The Washington Post, Feb.

Related Publications

Preprint
Full-text available
Suppose that we have $n$ agents and $n$ items which lie in a shared metric space. We would like to match the agents to items such that the total distance from agents to their matched items is as small as possible. However, instead of having direct access to distances in the metric, we only have each agent's ranking of the items in order of distance...
Article
A new index reflecting the extent to which a general theory is consistent is given in n-valued Łukasiewicz logical system based on Hausdorff metric and the truth degrees of formulas. Sufficient and necessary conditions for theories being consistent or inconsistent are given respectively.
Article
For more than two thousand years, mathematicians and other people believed in the "truth" of the Euclidean parallel axiom, before Lobachevsky and Bolyai about 1825-1830 found that it is possible to get a new geometry by substituting this axiom with the following.
Got a technical question?
Get high-quality answers from experts.