#picture_theory — Public Fediverse posts

For folks interested in #picture_theory, an intriguing question might be to ask “What colour is the pictorial depiction of the grass?”

From: @anvit
https://pixelfed.social/p/anvit/784823816000784808

For folks wondering why this mechanism is pertinent to #picture_theory, it is because the photographic nature of pictorial exposure is along this offset dimension.

As such, understanding the mechanisms behind the ratios and thresholds is crucially important when designing picture formation algorithms.

(Sample here from @barselino)

speculative #picture_theory that would possibly allow for research into more optimized algorithmic pictorial formations.

To break down the problem as simply as possible, we loosely have three problems to “solve”:
1. What is the underlying mechanism that drives the “derp goes white as mah exposure derp goes up”?
2. What is the *rate of change* that governs this mechanism?
3. What about “hue”?

All three of these incredibly deep problem surfaces are at least within striking distance under the +

When we think about #picture_theory, we can apply these mechanics to pondering what a more ideal pictorial representation within a medium might look like when we think about “algorithmic painters”.

A physical painter is engaged in the closed loop of painting value / tone in meatspace, and can paint their way toward more “ideal” results when we contrast against pictorial depictions formed from chemical film for example.

In chemical film, we *do* get some degree of axial chromaticty swings +

Important implications for #picture_theory. Sometimes when I am crawling through references in response to a question, I re-read the papers and find things that I had originally overlooked because of papers I had read *after* that paper. Re-reading is hugely informative.

I **very strongly** believe the “filmic” “brighter derp goes white” mechanism is enmeshed in the notions around cognitive fission / scission. The path to *better* than film pictorial formations rests in this tomb.

For those wondering why the “transparency” mode of visual cognition is worthy of some inspection with respect to #picture_theory, I would suggest that the “modes” of visual cognition are incredibly important as we imbue our pictures with meaning.

Not the higher order meaning, but rather the lowest possible order of meaning in how we encode the forms themselves. What makes a “line” a “line” or a “form” a “form”.

+

One might be wondering why I’ve been circling over this “transparency mode” in terms of a key visual cognition mechanic when it seems disconnected from #picture_theory.

It *can* seem disconnected up to the point that we realize that “tone mapping”, even when done poorly based on janky models or mechanics, is in fact *creating* an absolutely critical stimuli relationship in pictorial depictions. Without this additive energy, pictorial depictions quite literally tear apart.

For example, +

Somewhat interestingly, we are now in a position to describe accurately what the key difference is in #picture_theory Picture Formations between creative chemical film and the not-yet-a-picture of an electronic sensor.

An electronic sensor will only ever record scalar gains of what is in front of a camera. Colourimetry formed from these gains, or from an RGB render, is fundamentally devoid of the offset mechanism.

That is, without properly forming the “grammar” of a picture text, and properly+

Revising the thinking a little bit, given that each of the “offset” and “gain” dimensions can be considered in an increment *or* decrement orientation. There’s an interesting multi-stability baked into this sort of framing that is somewhat congruent with some of our visual cognition cue assemblies and how we reify / resolve these foundational ambiguities. #picture_theory

Happy holidays, folks!

(#picture_theory: Notice how the “atmospherics” (scare quotes, because this is a cognitive conclusion derived from our limited sensory orientation of the picture) make the more pure values scise to an assemblage based on the purity?)

I’m growing ever more convinced that the complex purity dynamics in pictures, specifically the slippery “thresholds” within a field, is a lower order neurophysiological signal relationship.

The whole “film-like roll off”, which *is not* “roll off” as that implies a “start” point where purity attenuates as opposed to an entire continuum, could plausibly be related to how we decompose a picture into “layered” meaning. #picture_theory

From: @petrikas
https://mastodon.art/@petrikas/111341128993768781

More on the idea that anyone peddling any sort of “Uniform Colour Space” or “Colour Appearance Model” based on a three dimensional system is out to lunch and lying…

Colour is a *cognitive* process, anchored in a purely differential substrate. As such, we cannot discuss colour without discussing the cognitive mechanisms and specifically, the differential relationships.

Here I’ve used the “Abney Effect” to demonstrate how contextually sensitive the stimulus is.

#picture_theory #colour_theory

Folks… we all understand that pictures of stuff isn’t the stimulus of standing there, right? If we don’t all start talking about this fact about #picture_theory, we will end up with more scientism of this sort of nonsense garbage.

“A Preliminary Investigation Into Individual Differences that Predict Men’s Preferences for Cues to Fertility in Women’s Faces”

Garbage scientism.

https://link.springer.com/article/10.1007/s40750-023-00223-0

- cognitively *mandatory* in pictures, we must be incredibly skeptical of this mechanic as being “natural”. Quite the contrary, there *is no general mechanic that yields this whiteness layering decomposition* in generic ecological cognition.

The whiteness decomposition *only occurs in pictures*, and predates photography as best as I can tell, to painting.

#picture_theory

Bit of a trippy paper for folks interested in #picture_theory perhaps. The idea of “rotating” memory encodings is rather fascinating, given it could be applied to picture cognition, as the Dalmatian in the Snow picture below.

From: @ekmiller
https://fediscience.org/@ekmiller/110978408574565219

- For folks wondering what this has to do with #picture_theory, we can accumulate various research on the On / Off and Off / On asymmetries^1 and get a sense that our notion of “gradients” and “sweep patterns” is largely moored in orthodoxy that is a tad deceptive. That is, instead of a “continuous” sweep, we should likely be focusing on differential field relationships.

We know the “down” direction is “more sensitive” than the “up” direction. +

From: @albertcardona
https://mathstodon.xyz/@albertcardona/110854630627236482

Some doozies in this paper for #picture_theory folks:

> The learning process led to progressive decorrelations in initially similar hippocampal neural activity within and across tracks, ultimately resulting in orthogonalized representations resembling a state machine capturing the inherent structure of the task.

+

#picture_theory

#picture_theory

(Note that the idea of field disparities, and signal disparities, ties into a potential model for understanding #picture_theory, but I’ll leave the above unlisted discourse as is, for anyone foolish enough to care what my opinion is.)

All I know at this point is that I lean away from the Newtonian directional stimulus to cognition pattern that plagues understanding. A reciprocating cognition loop seems far more reasonable, and visual field reification of things like picture-texts and literary-texts are most certainly leaning on the higher orders of visual cognition, away from the nonsense narrative of stimulus causation. Too many examples of instability to make the directionality “unidirectional”. #picture_theory

Note that the articulation arrangement of lower order spatiotemporal magnitudes, along with the middle order amplification and attenuation, leads to signals that create wildly different reified meanings in our visual cognition systems. We *read* and *cognize* the amplified and attenuated signals in our cognition.

These relationships are *incredibly* important when we form pictures out of camera data. No 3D model can afford any such “appearance modelling”. Nonsense!

#picture_theory

It is perhaps worth remembering that we, under no circumstances, can “turn this off”. All of our complex visual cognition apparatus is *always on* and *always cognizing*.

Every single time we look at a picture text, our “mode” of cognizing engages. We read the picture texts on many different cognitive levels.

#picture_theory

Eary putting in some hard work and willing to engage the politic. Might see AgX take shape and land in Blender trunk. A baby step perhaps toward more thinking about picture making, and wider thinking around the incredible complexity within conveyance of pictures. https://projects.blender.org/blender/blender/pulls/106355

#picture_theory

Should be absolute mandatory reading for anyone interested in image formation, and working our way toward more properly and well engineered picture formation chains. #picture_theory

These little seductive missteps are all over the place. When we discuss picture texts, we ought to be incredibly aware that we aren’t “seeing”, but “reading”, and account for the myriad of complexities involved in the visual cognition assembling by the authorship, and the subsequent interpretation by the readers.

#picture_theory

Note that the broad category of dorsal vision would likely correspond to the semiotics aspect of picture-text reading.

#picture_theory

I manually recreated the upper left demonstration of Adelson's Snake for folks to experiment with. I exported as SVG which *should* allow for automating the colour substitutions etc. https://editsvgcode.com/e4bj6znonsaleu8kc54

#picture_theory

Note that this is based on work from @jedsmith and @Jpzambrano. I intend to try and clean up the library file to use generic colourimetric functions and such, but it works for now, and it seems my PMs and DMs have been going a little bonkers for folks who want to try it.

#picture_theory

- The weirdness of the mathematical construct of a metameric black is that it seems entirely peculiar at first, but when we shift our lensing of vision away from a Cartesian understanding, toward a polar representation of opponency, the idea of a “counter-signal” becomes less bizarre.

#picture_theory

Was **not** expecting this quite exactly...

#picture_theory