How to select the type and style?

To simplify the tedious setup of your image generation, VToDoc provides Types and Styles to assist you.

1. Types: A type is a combination of style and image ratio, e.g. "Portrait" is a combination of "Extra-Photography" and "1024x1024", "Anime" is a combination of "SAI-Anime", "Extra-Semi-Real", "Extra-Great" and "896x1152". Of course, you can set it to a different value.
2. Styles:VToDoc offers more than 270 styles, which you can freely combine, up to 10 combinations. It should be noted that the prompt word and style need to match, for example, if you select "Terragen" and the prompt word is mainly a character description, the quality of the generated image is not ideal.

How to provide a good prompt word?

If you want to produce a high-quality image that meets your vision, in addition to choosing the Styles you need, you also need a good detailed and specific prompt. For example, if you only provide "cats", the resulting images are very extensive. But if you can narrow it down, for example, to the breed, color, or even pose of the cat, the resulting image will be closer to what you expected.

So how do you write a detailed and specific prompts? A good prompts should have the following elements:

1. Subject: Clearly indicate the main content of the image you want to generate, such as animals, plants, or other items. Including the subject's movements, expressions, eye states, clothing, decorations, etc., to enrich the characteristics of the subject.
2. Environment: Describe the environment in which the subject is located, including indoor or outdoor, lighting conditions, etc., to help the model better understand the desired situation.
3. Composition: Including the scene, shooting angle, depth of field, etc., to control the composition and perspective of the picture.
4. Lighting Effects: Specify the desired lighting effects, such as side lighting, backlighting, ambient lighting, etc., to enhance the mood and visual effect of the picture.
5. Color Atmosphere: Provide appropriate color keywords to change the tone of the whole picture.

By combining these elements, you can write a detailed and specific prompt that can more precisely guide the model to produce a picture that matches your vision.

Examples

You can consider the following elements when writing your prompts, but not all of them you have to include.

1. Subject

The subject is the main thing you want to present in your image. For example, if you want to generate a picture of a girl and type "a girl", it will be too concise, and the girl's posture, face, dress, background, etc. are very extensive, and the content of the generated image will vary greatly.

Therefore, in order to produce the expected picture more accurately, we need to provide a more specific description. For example:
Prompts: A beautiful girl with glamorous blonde hair, standing, smiling, and wearing a flowing green dress with intricate details

2. Environment

The environment to which the subject belongs also needs to be described. Place, time, environment, weather, lighting conditions, etc. can all be input prompts to affect the final background. Of course, it not only affects the background, but also the subject changes with the background information.
Prompts: A beautiful girl with glamorous blonde hair and blue eyes, standing by a cherry blossom tree, smiling, wearing a flowing green dress with intricate details, sunny day

3. Composition

The composition includes the scene, shooting angle, depth of field, etc.
For example, "panorama" can show the whole of the subject, and for the range of the body including head to toe, it is more commonly expressed as "whole body". In the same way, there is also the "upper body", that is, from the waist up.
Prompts: A beautiful girl with glamorous blonde hair and blue eyes, standing by a cherry blossom tree, full body, high heels on her feet, smiling, wearing a flowing green dress with intricate details, sunny day

4. Lighting Effects

Specify the desired lighting effects, such as side lighting, backlighting, ambient lighting, etc., to enhance the mood and visual effect of the picture.
Prompts: A beautiful girl with glamorous blonde hair and blue eyes, standing by a cherry blossom tree, full body, high heels on her feet, smiling, wearing a flowing green dress with intricate details, sunny day, warm sun, cinematic lights, high contrast

5. Color Atmosphere

The whole picture is greenish, and you can add more colors to it, such as gold.
Prompts: A beautiful girl with glamorous blonde hair and blue eyes, standing by a cherry blossom tree, full body, high heels on her feet, smiling, wearing a flowing green dress with intricate details, bright golden atmosphere, sunny day, warm sun, cinematic lights, high contrast

Other than that:

The closer the keyword is to the front of the prompt word, the higher the weight, which means that the higher the position, the more pronounced the keyword will be in the generated image. The "whole body" is earlier than in this example.