Super skin texture enhancement | Ultra-clear pore details 1
Model Information
Super Skin Texture Enhancement | Ultra Clear Pore Details
This model focuses on the refined presentation of skin textures, performing well in portrait generation and skin detail optimization scenarios. Its key feature is the ability to showcase skin textures extremely delicately, clearly presenting pores, wrinkles, and subtle skin undulations, giving portrait photos a hyper-realistic texture. It helps creators achieve realistic skin detail when generating character images, making it particularly suitable for creations pursuing ultimate realism, simulating realistic skin textures in photography, and showcasing beauty products, highlighting skin details as a standout feature in the image.
No trigger words.
Recommended LoRa weight: Suggested weight around 0.5.
Sampling method: Euler.
Iteration steps: 25 to 35 steps.
Recommended dimensions: 1024×1536, 1536×1024 and above.
Prompt guidance coefficient (CFG): 3.5 to 4.5.
High-resolution fix: ESRGAN_4x.
Repainting amplitude: 0.3 to 0.4.
Recommended tags
This model responds better to skin-related tags, such as “ultra detailed skin texture,” “clear pores,” “enhanced skin micro details,” “smooth skin,” “matte skin,” “glowing skin,” “blemished skin,” etc. It can precisely combine these tags to generate skin details in various states and also provides positive feedback for tags like “human portrait,” “close up,” etc., facilitating the emphasis on showcasing skin textures.
Super Skin Texture Enhancement | Ultra Clear Pore Details
This model focuses on the refined presentation of skin textures, performing well in portrait generation and skin detail optimization scenarios. Its key feature is the ability to showcase skin textures extremely delicately, clearly presenting pores, wrinkles, and subtle skin undulations, giving portrait photos a hyper-realistic texture. It helps creators achieve realistic skin detail when generating character images, making it particularly suitable for creations pursuing ultimate realism, simulating realistic skin textures in photography, and showcasing beauty products, highlighting skin details as a standout feature in the image.
No trigger words.
Recommended LoRa weight: Suggested weight around 0.5.
Sampling method: Euler.
Iteration steps: 25 to 35 steps.
Recommended dimensions: 1024×1536, 1536×1024 and above.
Prompt guidance coefficient (CFG): 3.5 to 4.5.
High-resolution fix: ESRGAN_4x.
Repainting amplitude: 0.3 to 0.4.
Recommended tags
This model responds better to skin-related tags, such as “ultra detailed skin texture,” “clear pores,” “enhanced skin micro details,” “smooth skin,” “matte skin,” “glowing skin,” “blemished skin,” etc. It can precisely combine these tags to generate skin details in various states and also provides positive feedback for tags like “human portrait,” “close up,” etc., facilitating the emphasis on showcasing skin textures.