Dream Mirror: Tricolor Daydream

0
0
0
0
CGI styleAnime StyleGirlSpace & SceneStyle Boost
Baru Diperbarui: Diterbitkan Pertama:
CGI style,Anime Style,Girl,LoRA,SD 1.5Image info
CGI style,Anime Style,Girl,LoRA,SD 1.5Image info
CGI style,Anime Style,Girl,LoRA,SD 1.5Image info
CGI style,Anime Style,Girl,LoRA,SD 1.5Image info
CGI style,Anime Style,Girl,LoRA,SD 1.5Image info
CGI style,Anime Style,Girl,LoRA,SD 1.5Image info
CGI style,Anime Style,Girl,LoRA,SD 1.5Image info
CGI style,Anime Style,Girl,LoRA,SD 1.5Image info
CGI style,Anime Style,Girl,LoRA,SD 1.5Image info
CGI style,Anime Style,Girl,LoRA,SD 1.5Image info
CGI style,Anime Style,Girl,LoRA,SD 1.5Image info
CGI style,Anime Style,Girl,LoRA,SD 1.5Image info

Tri-color Daydream is a set of red, yellow, and blue stylized works created by myself at Midjourney. The series is mainly based on Chinese fantasy themes, imagining contemporary and ancient, future coexistence in a super-realistic illustration form, with dream and reality overlapping in the imagery space.

Dataset Overview


I have always had a study demand for the new Chinese visual culture, and this series can also be seen as a phase achievement. It integrates contemporary art, especially surrealism, and the aesthetic features of Chinese comics and illustrations. How to extract the aesthetic features of contemporary Chinese art and humanities data that can represent (or partially represent) using the Lora model. Although this model is not a perfect match for the expected result, after several evaluations, especially in combination with other styles of lora used by the author, it produces many interesting and intriguing images.

This series excels in bizarre fantasy scenes, which can be well combined with character narratives, mainly by refining aesthetic features through the three primary colors.

I especially recommend the dark hellish style (  https://www.shakker.ai/modelinfo/2b9bd5d0366b4c58b845a30a64feddbc  ) with a recommended weight between 0.2-0.4, not too high.


The cover image effect is also greatly assisted by this series, details can be found in the example images posted at the top of the return image area (including parameters).

Weight Recommendation: 0.6-0.9

Base Model Recommendation: Dream Mirror China_v1.10.2, dreamshaper 6BakedVae, ghostmix_v20

Sampler Recommendation: DDIM (imaginative but may have logical issues), DPM++ 2M Karras, DPM++ 3M SDE Exponential

(the latter two generally have good realism and visual logic in the generated images)

Trigger Word: CNDaydream (optional)


Thank you for reading! Let's exchange more image returns~


Disclaimer:

In principle, this model is only for communication and learning purposes and should not be used for commercial purposes.All content generated based on this model is the user's independent creative behavior after their own judgment. Please comply with relevant laws and regulations.Any intellectual property issues caused by using this model to generate content are the user's own responsibility.Without the author's written consent, this model service should not be integrated into any third-party software or service, and should not be disclosed to any third party in any way that directly or indirectly implies a relationship between the generated content and this service. Violation of this clause may result in the author seeking compensation for losses incurred.

Diskusi

Paling Populer
|
Terbaru
Kirim
Segera Hadir
Unduh
(0.00KB)
Deskripsi Fungsi
Tipe
Hitungan Hasil
0
Unduhan
0
Parameter yang Direkomendasikan
Kombinasi yang DirekomendasikanCheckpoint
Tidak Ada
Berat yang direkomendasikan
0.8
CFG
7
VAE
Tidak Ada

Galeri

Paling Populer
|
Terbaru