welfare machine
Exploring the selfie style tips of internet celebrities
I have been exploring the trending selfie style on different platforms these days.
[AI painting]
Related links: This article's
Experience it now.
In the past few days, the overwhelming trend of welfare girls has been centered around low-quality selfies.
For example: lens flare, dim environments, stains on mirrors, cheap and tacky outfits, chaotic and blurry backgrounds, etc.…
In short, it is to use defects to neutralize the excessive aesthetic of AI to increase the realism of the image.
The reason why the LLM series is better is that it easily meets requirements through the keyword 'mediocre', while traditional diffusion systems like MJ need to use precise prompts to lower aesthetics, making it relatively more difficult.
Midjourney
 |  |
|
|
 |  |
|
|
FLUX(Krea Optimized)
Krea's built-in Flux scale is much wider than MJ's; although there is some review, some borderline cases are still quite easy to achieve.
 |  |
|
|
 |  |
Chinese coser, athletic, slightly muscled, taking TikTok selfie in a seaside assorted wooden room through the dusted mirror, you can see the beach through the window, low angle, lolicore, breakcore, cute face, full body, white bikini, webcam, iPhone 13, low quality photo, active pose, plain aesthetic, webcam, background movement blurry, slightly dusted unclear mirror glass, clean but messy room, low angle view, focus on the leggings as product advertisement, cartoon phone shell | Chinese coser, taking TikTok selfie in her own room through the dusted mirror, low angle, lolicore, breakcore, cute face, full body, unbuttoned white shirt, short pink jeans hot pants, webcam, iPhone 13, low quality photo, plain aesthetic, webcam, background movement blurry, slightly dusted unclear mirror glass, clean but messy room, low angle view, focus on the leggings as product advertisement, cartoon phone shell |
Doubao
Relatively speaking, Doubao, as an LLM-style drawing, can understand prompts very well and describe requirements in a more direct and detailed way, so the resulting images are much more precise than ordinary diffusion models.
 |  |
|
|
Related links: This article's
Experience it now.
