HOME04/ WHY DO I FEEL PRETTIER?> METHOD AND ANALYSIS

04/ WHY DO I FEEL PRETTIER?

METHOD AND ANALYSIS

RESEARCH QUESTION

In the discussions on the effective harmfulness of the types of alterations used in beautification filters, the analysis wanted to investigate the issue by analyzing the type of alterations present among Perfectionist beauty filters. Since this aesthetics aimed at improving the appearance, the research wanted to verify if actually the types of changes can be associated with plastic surgery practices becoming concretely harmful.

DATA COLLECTION

The data collection in this case was done differently than in previous research protocols: if in the previous cases the data analyzed were Instagram posts, in this instance the data analyzed was the filter itself, applied to a unique face. Instagram profiles that sposorise AR filters were used as starting points to find popular creators and effects. Through the profiles of Perfectionis Aesthetics (@aestheticfilters @beautyfilters), 27 creators were collected:

@alexandra.kisa @aaaneka @daryabiller @erikavelez @etolitvinova @exotic.cancer @filtersbeauty @filterscreative @hellyluv @itskovalevskaya @ksubones @malinkamau @patchisphotos @sasha_soul_Art @sasha. ray @sophie @stasmileev @thakatetrump @top.beauty_ @filtersbeauty @vse maski @vvsnaleto @whoisfrancisco @why.not.say @xeniabelskaya @yulya.kors @zemtsova_lubov, for a total of 320 filters collected.

IMAGES ANALYSIS AND PROCESS

The final images to be analyzed has been created through a process of application of the filter on a non-existing face of thispersondoesnotexist.com. The face has been printed and through the camera of an Iphone 8 (positioned at the same distance and position) 320 photos of the face has been taken with different effects applied. Every single shot was saved and renamed with the actual title of the effect.

At least, the images has been cropped to the same size through the Photoshop batch. To create a grid of images placed in color gradient from lightest to darkest, they were processed using image J to calculate the average brightness of each photo. The miscalculations were fixed manually by repositioning the image in the proper position according to the color scale.