初始圖片

1.GUIDANCE SCALE
Higher guidance will make your image closer to your prompt.
(指導範圍:更高的指導範圍值將使您的圖像更接近您的提示。)
2.NUM INFERENCE STEPS
More steps will result in a high quality image but will take longer.
(推理步驟數:更多的步驟將產生高質量的圖像,但會花費更長的時間。)
3.PROMPT STRENGTH
Higher guidance will make your image closer to your prompt. 1 corresponds to full destruction of information in init image
(即時強度:更高的指導將使您的圖像更接近您的提示。注意:1對應於初始化鏡像中信息的完全銷毀)
以下為一些成果:







歡迎深入討論:
↩️ obsidian 黑曜石 line社群:https://hi.switchy.io/oblearn
↩️ heptabase line 社群:https://hi.switchy.io/heptalinbgroup
↩️ ChatGPT line 社群: https://hi.switchy.io/chatgptline