AICaptionHelper is an innovative tool designed to streamline the preparation of your datasets for Stable Diffusion checkpoints, LoRA’s etc. Our application empowers you to generate precise text caption files for each image using an intuitive UI and combining your insights with optional AI assistance for enhanced language quality.
While AICaptionHelper harnesses the advanced language model of OpenAI, the importance of your input remains paramount. You'll guide the captioning process by providing specific details about your image – from the subject and their hairstyle to their activities, locations, and facial expressions. Optionally, you can input your OpenAI API key, allowing the AI to refine your input, ensuring impeccable spelling, grammar, and a more natural flow of language.
Collaborative Captioning: Combine your expertise with optional AI-powered assistance to craft comprehensive image captions.
Optional OpenAI Integration: Utilize your OpenAI API key to refine captions, improving grammar, spelling, and linguistic fluidity.
Tailored for Stable Diffusion Models and LoRA’s: Optimized for creating descriptive datasets for Stable Diffusion Models and LoRA’s.
User Input Fields: Provide detailed information about your images across various categories, driving the caption creation process.
Caption Preview and Editing: Review and refine the AI-enhanced caption before final submission.
Batch Processing: Effortlessly load multiple images from a chosen directory and advance to the next image after each caption is generated.
This is a work in progress. Please provide feedback.
Please read the README file and watch the video walkthrough.
V2.1 fixes: you can now update the text box with any changes you want and when you press "caption it" it will save that text to the file.