Image posted by roct2468836

txt2img
Text-to-image generation. This technique uses natural language prompts to generate images from scratch using AI models like Stable Diffusion or DALL·E.
img2img
Image-to-image transformation. This technique modifies an existing image based on a prompt or style, preserving composition while changing content or quality.
inpainting
AI-based image editing. It allows the user to erase or mask parts of an image and regenerate them using a prompt, often used for fixing or modifying specific regions.
workflow
A structured process of using multiple AI tools or steps (e.g., txt2img → img2img → upscaling) to refine and generate images systematically.
vid2vid
Video-to-video synthesis. This technique transforms an existing video into a new one, often stylizing or altering it using AI.
txt2vid
Text-to-video generation. It generates video clips from text prompts using models like Sora or Runway.
img2vid
Image-to-video animation. It takes still images and animates them into moving content, often with inferred motion or transitions based on context.
controlnet
A powerful extension for AI models that allows control over image generation using poses, depth maps, edges, etc., to ensure structural accuracy and consistency.

Generation data

COPY ALL

Resources used

WAI-illustrious-SDXL
Checkpoint
v14.0
girl
LoRA
V1

Prompt

External Generator

txt2img

cat ears, cat tail, white hair, golden eyes, school uniform, thighhighs, detailed hands, sharp eyes, masterpiece, high quality, best quality, 1girl, animal ears, looking at

Other metadata

cfgScale:7.5

steps:30

sampler:DDIM

Discussion