All Collections
Styles
When creating styles, how do categories make a difference?
When creating styles, how do categories make a difference?
Updated over a week ago

Categorization helps in organizing and defining different styles, which allows for more efficient and effective training. Once categories are made, narrowing the focus of the training process allows the model to concentrate on learning distinct characteristics and patterns, leading to a more accurate style.

At starryai, we offer 5 categories:

  • Art: Tailored for abstract and creative visual interpretations.

  • Portrait: Optimized for capturing facial features and expressions.

  • Characters: Designed for fictional personas and consistent characters.

  • Illustration: Suited for detailed and imaginative artworks.

  • Photography: Fine-tuned for photorealistic high-quality images.

💡Tip: For the best outcomes, ensure your dataset images match the chosen category since the base model is pre-tuned to that style. Using mismatched styles, such as arty images for a photography category, can affect the fine-tuning and influence the results.

Below are examples of the unique images that can be produced using different categories.

Photography:

Art:

Illustrations:

Portrait:

Characters:

Did this answer your question?