is it red? what is red? [defining zombie pupils]

Bastard Gan Punks are GAN generated, and GAN products are more complex than we tend to think. Great minds in the community come together and discuss traits: what are these particular pixels here? how did they come to be? what did the GAN intend here? is it clean? is it red? These analysis/debates are so fundamental because a consensus is required to move forward with a final decision/definition of what these groups of pixels are pointing at. More often that not, trait = value. It is important to emphasize that traits are hashtags and because NFT valuation nowadays is still largely dependent on traits, hashtags alas are still important. More often than not, a holder proposes a change/correction of a trait and they have to defend their thesis successfully until the DAO is convinced and the proposal passes (or not). This particular activity is a great exercise for the holder and for the DAO because it encourages trait discussions, trait hunts, corrections and frankly makes a connoisseur out of all of us. Art is a fundamental human right and an evolution/change in art/trait interpretations must always remain possible. Defining red One of the hottest debates still occurring in the BastardDAO concerns the zombie red definition. According to the current trait constitution pioneered by trait scientist Snooplyin: 1.1.4 Zombie

1. If at least one red pupil, then zombie. See OG punk zombie example.
2. Pupil looks back against direction facing.
3. Red dominant, not orange, pink, brown, purple, etc.
4. 4-pixel glasses lenses: If the pupil pixel is red and distinct from the rest of the glasses lens/eye patch, even if just different shades of red, then it counts as a zombie.
5. 9-pixel glasses lenses: If the lens is made of different primary colors and pupil is distinct from surrounding pixels.
   1. Ignore color variation if the lens is different shades of the same primary color.

In this definition above, part 3 indicates that the pupil must be [red dominant]. In color spaces, red dominance can indicate several things: for example, in the RGB space, it could indicate high red values with less input from green or blue. In other spaces it could indicate red hue + lightness or red values + saturation, etc... The origin or explanation of the wording above in the constitution, would be red dominant = according to our eyes. Yet, our judgement for color dominance differ because 1- humans do not see or define colors the same way just by perception 2- monitor colors differ  this leads to a very subjective interpretation of red dominance which in turn leads to conflict. That conflict will be renewed every time there is a zombie proposal. Before I begin the proposal, I would like to shamelessly copy-paste info about the different color spaces and talk about the method chosen to define red. There exists many color spaces where colors are calculated and interpreted differently. Some are more robust than others, some are good for certain use cases others are not. The main color spaces are RGB, HSV and CIELab (or Lab), which of these three is more accurate and more representative? Turning something abstract like color spectrum into color models is incredible! RGB: Most digital images are stored as RGB images, and have to be converted to other color spaces. A color in RGB space is specified with a red, green, and blue coordinate, and each channel is 8-bit, meaning it has 28=256 possible values. For simplicity, we can bound each of these channels between 0 and 1, with 0 being the no contribution from that color channel, and 1 being the maximum. So pure blue, for example, has an RGB triplet of red = 0, green = 0, and blue = 1, or [0, 0, 1]. Yellow is [1, 1, 0], meaning that blue and yellow are on opposite ends of RGB color space. RGB space, however, takes the pixels in a digital image completely at face value, and makes no attempt to correct for the variables described above – namely, the available light in the environment or the sensitivity of the camera. This is fine if all the images in your dataset were taken under identical lighting conditions with the same camera, and those lighting conditions adequately mimic the relevant lighting conditions for the research question, but this isn’t always necessarily the case. Another major criticism of RGB space is that it is device-dependent, meaning that the same RGB triplet will be displayed slightly differently on different monitors. HSV: Like RGB, hue-saturation-value color space relies on three channels, all of which range from 0 to 1. But while RGB color space is modeled on the peak color sensitivities of the human eye, HSV is modeled on the ways that people consciously break down colors. You probably don’t think of sunflower yellow as equal parts red-cone and green-cone stimulation, for example, but you think of it as a bright, saturated yellow color, whereas beige or tan would be desaturated pale yellows. HSV color space tends to be fast and intuitive, but it’s also a poor proxy for color complexity as perceived by most organisms. This doesn’t necessarily make it irrelevant for analyses, especially if your goal is to quantify image similarity or dissimilarity for some reason other than trying to mimic organismal color perception, but it does make this color space a relatively unpopular choice in biological research. CIElab: Of the spaces that colordistance can use, CIELab (or CIELAB, or CIE Lab) color space is both the most complex and the most robust for quantitative comparisons. CIELab space is defined by the International Commission on Illumination (CIE) with the intent of being a perceptually uniform color space, meaning that sets of colors separated by the same distance in CIELab space will seem about equally different. The three channels of Lab space are luminance (black to white), a (green to red) and b (blue to yellow). This is a convenient and fairly intuitive way to organize a color space, but because the boundaries human color vision don’t fall neatly within a perfect cube, neither does CIELab. When working with digital images, CIELab is often your best option. When working with digital images, CIELab is often your best option. (Ref: https://cran.r-project.org/web/packages/colordistance/vignettes/color-spaces.html) In a nutshell, CIElab is the most advanced color space and the most accurate in the context of human vision. It also performs better than RGB and HSV (ref: https://ieeexplore.ieee.org/document/5697476). From this moment onwards, I will use the CIELab system (from here on I will call it Lab) as my standard for analyzing zombies and zombie hybrids pupils. To define red dominance, a robust formula called [DeltaE] has been employed and it uses the sample Lab coordinates to measure the distance from OG cryptopunks zombie red lab coordinates. The closer the sample is to the OG reference, the lower the value. 0 being the lowest value and 100 the highest. In this proposal we will vote on the distance or DeltaE range that we deem close to red to call it [the range of red dominance]. Defining DeltaE: The CIE 2000 color difference formula was developed to solve the problem of the differences in the evaluation between color meters and the human eye. The problem is caused by the difference in the shape and size of the color discrimination threshold of the human eye. The CIE 2000 color difference formula is not an attempt to build a color space. Instead, it defines a calculation so that the color difference calculated by color meters becomes close to the color discrimination threshold of the human eye on the solid color space of CIE Lab (Lab* color space). Specifically, weight is assigned to the lightness difference ΔL*, saturation difference ΔC*, and hue difference ΔH’ by using weighting coefficients SL, SC, and Sh respectively. These weighting coefficients SL, SC, and Sh include the effects of lightness L*, saturation C*, and hue angle H. Consequently, the calculation incorporates the characteristics of the color discrimination threshold of the human eye on the color space of CIE Lab (Lab* color system): 1) Saturation dependency, 2) Hue dependency, and 3) Lightness dependency. Delta E value Meaning 0 - 1 A normally invisible difference 1 - 2 Very small difference, only obvious to a trained eye 2 - 3.5 Medium difference, also obvious to an untrained eye 3.5 - 5 An obvious difference

6 A very obvious difference Another explanation: (ref: http://zschuessler.github.io/DeltaE/learn/) Delta E Perception <= 1.0 Not perceptible by human eyes. 1 - 2 Perceptible through close observation. 2 - 10 Perceptible at a glance. 11 - 49 Colors are more similar than opposite 100 Colors are exact opposite.

OG cryptopunks zombie pupil (L,a,b) coordinates: 53.03, 79.87, 67.02 OG CP zombie horned rim glasses pupil (L,a,b) coordinates: 59.36, 52.06, 19.79 I used these coordinates as a reference for calculating the totality of zombie and zombie hybrid pupils. Tools used: For converting RGB values to Lab values: https://colorizer.org/ For calculating the DeltaE (using CIE2000 formula): http://www.brucelindbloom.com/index.html?ColorDifferenceCalc.html Pinetools for collecting values from each zombie: https://pinetools.com/image-color-picker Note: I used background-less BganPunks zombies and zombie hybrids kindly provided by Witrebel to eliminate the color influence from the glicpixxx background since some pixels are not 100% opaque. Please find attached to this proposal a images of the values for each zombie or zombie hybrid and three charts of the data above. My calculations show that around 60% percent of the total zombies (incl. hybrids) reside in the distance from 0 to 15 and 75% from 0 to 20. The lowest value being 0.58 and the highest value 41.88. To visualize what colors represent 6-10-15-20-25-30-25-40 and 45, I added two examples for each range, so that the DAO can judge what these numbers mean (images attached). Now that each ID has a value and visual examples are provided, the moment has come for the DAO to define an acceptable range. The proposal will run for 15 days and will be a weighted voting, please do not hesitate to ask me questions. CMFB