=====================================================

Update v3.0

Version 3.0 optimizes some materials, adds a large number of high-quality deep image materials, and improves the generalization performance of the model to some extent with thousands of materials. At the same time, the tags of each material are optimized and adjusted, which took a lot of time. The purpose is to enhance the accuracy of the prompts and improve the recognition level compared to before. Therefore, when generating images, try to describe the original image as accurately as possible. This can strengthen the stylistic effect of the generation.

Image quality control prompts:

Positive words: depthstyle, masterpiece, best quality, monochrome, grayscale, black background, simple background

Negative words: lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry

Fill in the content prompts according to your needs. You can refer to the example image

Recommended sampling steps: 20-30

Recommended sampling method: Euler a

Prompt word guiding coefficient (CFG Scale): 4-6

Suggested direct output image sizes: 512*512, 512*768, 768*512

Test more parameters according to your usage for effect evaluation.