Degenerate images

Running the “orderings” code on a larger selection of images from Guttorm Guttormgaard’s collection (still under construction), several images caused the code to break as “exceptions” (unexpected errors or situations occured). The first image caused the texture processor to throw a mysterious floating point exception.

34c_MG_8917WEB

In a later pass, the following image caused an exception because it was apparently the first image to have a single SIFT feature, and the code that reads the SIFT features from a file got confused and caused an error related to the size of the number of dimensions of the array.

72a_IMG_5979WEB

In mathematics & engineering, these kind of exceptional cases are sometimes referred to as “degenerate” cases.

In mathematics, a degenerate case is a limiting case in which an element of a class of objects is qualitatively different from the rest of the class and hence belongs to another, usually simpler, class. Degeneracy is the condition of being a degenerate case.
source

“One Millisecond Face Alignment with an Ensemble of Regression Trees”

This paper addresses the problem of Face Alignment for
a single image. We show how an ensemble of regression
trees can be used to estimate the face’s landmark positions
directly from a sparse subset of pixel intensities, achieving
super-realtime performance with high quality predictions.
We present a general framework based on gradient boosting
for learning an ensemble of regression trees that optimizes
the sum of square error loss and naturally handles missing
or partially labelled data. We show how using appropriate
priors exploiting the structure of image data helps with ef-
ficient feature selection. Different regularization strategies
and its importance to combat overfitting are also investi-
gated. In addition, we analyse the effect of the quantity of
training data on the accuracy of the predictions and explore
the effect of data augmentation using synthesized data.

From One Millisecond Face Alignment with an Ensemble of Regression Trees Vahid Kazemi and Josephine Sullivan; KTH, Royal Institute of Technology
Computer Vision and Active Perception Lab
Teknikringen 14, Stockholm, Sweden

Matthewearl_faceswap

Cited by this blog post which uses the technique, among other things to match and then project one face onto another.

Computational linguistics as seen by Stanislaw Lem

One day Trurl the constructor put together a machine that could create anything starting with n. When it was ready, he tried it out, ordering it to make needles, then nankeens and negligees, which it did, then nail the lot to narghiles filled with nepenthe and numerous other narcotics. The machine carried out his instructions to the letter. Still not completely sure of its ability, he had it produce, one after the other, nimbuses, noodles, nuclei, neutrons, naphtha, noses, nymphs, naiads, and natrium. ‘This last it could not do, and Trurl, considerably irritated, demanded an explanation.
“Never heard of it,” said the machine.
“What? But it’s only sodium. You know, the metal, the element…”
“Sodium starts with an s, and I work only in n.”
“But in Latin it’s natrium.”
“Look, old boy,” said the machine, “if I could do everything starting with n in every possible language, I’d be a Machine That Could Do Everything in the Whole Alphabet, since any item you care to mention undoubtedly starts with n in one foreign language or another. It’s not that easy. I can’t go beyond what you programmed. So no sodium.”

Read more

How the world was saved, in The Cyberiad, Stanislaw Lem.

Anarchive, Wolfgang Ernst

“The term archive has been the dominant metaphor for all kinds of memory.”
Listen online

An interview made by SON[I]A.

Scandinavia is a joke

[caption id="attachment_347" align="alignnone" width="296"]The Scandinavian Institute of Comparative Vandalism at work. Taken by Franceschi on Gotland in 1964, the photo shows Jacqueline de Jong, Asger Jorn, Ulrik Ross (on the roof of the Citroën) and an uidentified fourth man. The Scandinavian Institute of Comparative Vandalism at work. Taken by Franceschi on Gotland in 1964, the photo shows Jacqueline de Jong, Asger Jorn, Ulrik Ross (on the roof of the Citroën) and an uidentified fourth man.[/caption]

Ellef Prestsæter: Did you feel that you were collecting evidence for a specific archaeological argument?

Jacqueline de Jong: Yes, sure! It was the argument that there had been people, like the Visigoths, travelling all over the globe. Jorn wanted to show that the images are the same in Italy, Spain, Portugal, wherever you go.

(more…)

Francis Galton’s Composite Portraiture meets Wittgenstein’s camera

CriminalComposites

Galton also devised a technique called “composite portraiture” (produced by superimposing multiple photographic portraits of individuals’ faces registered on their eyes) to create an average face. In the 1990s, a hundred years after his discovery, much psychological research has examined the attractiveness of these faces, an aspect that Galton had remarked on in his original lecture. Others, including Sigmund Freud in his work on dreams, picked up Galton’s suggestion that these composites might represent a useful metaphor for an Ideal type or a concept of a “natural kind” (see Eleanor Rosch)—such as Jewish men, criminals, patients with tuberculosis, etc.—onto the same photographic plate, thereby yielding a blended whole, or “composite”, that he hoped could generalise the facial appearance of his subject into an “average” or “central type”. (Wikipedia)

On galton.org, one can find an impressive listof fac-similes of Galton’s articles concerning composites and analytical photography.

(more…)

250 000 labels

basket

This document, edited by Antonio Torralba, contains the notes written by Adela Barriuso describing her experience while using the LabelMe annotation tool. Mrs Barriuso has no training in computer vision. In 2007 she started to use LabelMe to systematically annotate the SUN database. The goal was to build a large database of images with all the objects within each image segmented and named.

By the way of Femke Snelting

(more…)

Agatha Christie smoking Asger Jorn’s cigar

Poctracingforprint

Ellef Prestsæter: Readers of Concreta, beware! … Engaging with you, Asger, in the context of this special issue on Vandalism and Iconoclasm calls for a strict demarcation between vandalism on the one hand and iconoclasm on the other. While the Scandinavian Institute of Comparative Vandalism (SICV) connects you intimately with the concept of vandalism you are anything but iconophobic. Your notorious détournements or modifications provide a case in point: there clearly is a brutally destructive moment in overpainting these tableaus bought at flea markets, but the point was never to annihilate or make fun of them. You were rather driven by a profound interest for the “popular” art form these pictures represented. The overpainting is a form of resurrection or remobilization, or, on a somewhat more prosaic note, a form of image analysis. And the endpoint was always a new image. Whatever you may have meant by “vandalism”, your practice was always rooted in an engagement with, and even a love for, images. The iconophilic vandalist? The notion carries a certain attraction …

Asger Jorn: An interest in dispersed attention, such as the one modern art has shown throughout this century, may contribute to the deterioration of culture’s authoritarian patent makers, and surely it is this practice you would call comparative vandalism.

(more…)

Black boxing conversion

[…]

Bernhard Siegert: In a way, you can describe the business of cultural techniques as the opening up of black boxes. If you think of concepts or even symbols as black boxes, when you open them up, what comes out are cultural techniques.

One of the things I have been studying recently is the very famous Macy Conferences. All these star thinkers there — Norbert Wiener, Margaret Mead, Gregory Bateson, Warren McCulloch — had no clear concept of what the digital was.

And a group of neurologists wants to describe the nervous system as a digital machine. That is what McCulloch and Walter Pitts already had in mind in the ’40s: They were interested in a concept of the digital that lies in the real, in the natural. They really thought that nerves, synapses, work digitally. And so there is great confusion.

(more…)

Capta

Etymologically the word data is derived from the Latin dare, meaning ‘to give’. In this sense, data are raw elements that can be abstracted from (given by) phenomena – measured and recorded in various ways. However, in general use, data refer to those elements that are taken; extracted through observations, computations, experiments, and record keeping (Borgman 2007). Technically, then, what we understand as data are actually capta (derived from the Latin capere, meaning ‘to take’); those units of data that have been selected and harvested from the sum of all potential data (Kitchin and Dodge 2011). As Jensen (1950: ix, cited in Becker 1952: 278) states:

it is an unfortunate accident of history that the term datum… rather than captum… should have come to symbolize the unit-phenomenon in science. For science deals, not with ‘that which has been given’ by nature to the scientist, but with ‘that which has been taken’ or selected from nature by the scientist in accordance with his purpose.

The data revolution, Rob Kitchin.