Controls & Features to Use in Prompts

While using Imagine 1.0, there are some key factors and controls to add to your prompt for ultra-realistic results:

1. Negative Prompts

Adding ‘what not to add’ in your prompt helps the AI image generator understand and avoid unnecessary elements. For instance, you can add “no text” or “no filters” to avoid visual artefacts in the generated image.

3. Descriptors

Without a clear description of the scene, environment, characters, and objects, achieving realism wouldn’t be a possibility. Make your prompt concise but comprehensively descriptive for contextual understanding. Include color palette, surroundings, qualities of the main subject, lighting, tone, ambiance, and more.

4. Framing & Styles

ImagineArt 1.0 allows you to add camera positions, angles, and styles to improve visual compositions. Add stylistic preferences such as “photorealistic” or “Vogue-style aesthetic” to tailor the generated image to your requirements.

5. Aspect Ratio & Variations

After typing in the prompt, select the right aspect ratio (1:1, 3:4, 4:3, 9:16, 16:9) and number of images from the options. You can generate up to 4 images per generation. Turn on the prompt enhancer if you want to further refine the entered text.

Prompt Templates

Copy, paste, and edit these templates as per your image generation requirements for super-realistic outputs. Each template fulfills a different creative need:

Template A: Human Portrait

A (subject) with (distinctive attribute), standing (or posed) in (environment), bathed in (lighting/mood), shot with an (lens/framing) — hyperrealistic, ultra-detailed, 4K. Negative: (what to exclude)

Example:

• “A young woman with freckles and auburn hair, standing in a misty forest at dawn, soft backlight illuminating dew, shallow depth of field with 85mm lens — hyperrealistic, ultra-detailed, 4K. Negative: no text, no other people.”

Generated by Imagine 1.0

Template B: Product Mockups

A (product/object) in (environment/setting), rendered with (material properties), under (lighting setup), with (style reference if any) — ultra-detailed, photorealistic. Negative: (exclude unwanted elements)

Example:

• “A matte black mechanical wristwatch on a polished marble slab, subtle reflections, rim lighting from the top, soft shadows, high contrast — photorealistic, ultra-detailed. Negative: no props, no distracting background.”

Generated by Imagine 1.0

Template C: Scenic Landscape

A sweeping (landscape/cityscape/fantasy scene) at (time of day), with (key features), in (style), with (lighting/atmosphere) — ultra-detailed, 4K. Negative: (if needed)

Example:

• “An empty city at twilight, neon signs reflecting on wet streets, abandoned vehicles parked on the streets, towering architecture draped in vines — cinematic, photorealistic, ultra-detailed, 4K. Negative: no people.”

Generated by Imagine 1.0

Prompt Iteration & Refinement

Crafting an ultra-realistic image on Imagine 1.0 might require prompt refinement and reiteration. Here’s how you can fine-tune your prompt for improved realism:

1. Start Simple

If it is your first time testing out the model, start with a simple prompt including a subject and basic setting details. See what Imagine 1.0 creates from that simple prompt and then reiterate. However, don’t overcomplicate or lengthen the prompt too much.

2. Add Elements & Customization

Once you understand the image generation capabilities, refine your prompt with visual details such as lightning, camera movements, visual style, textures, character appearance, and more. Add more details to your prompt gradually.

3. Negative Prompting

In case you are pretty clear about what you don’t want in your image, add a negative prompt to exclude those elements.

4. Variant Comparison

It is better to create a couple or 4 images per generation to compare the visual quality, clarity, and composition. This allows you to make the right selection while giving you an idea of how to further refine your prompt.

5. Reuse Base Prompt

In case you want to generate a sequence of images with the same subject and settings, make sure to reuse the base prompt with added customization to maintain character consistency and visual coherence.

6. Refactor the Prompt

In case the generated image doesn’t quite meet your vision, edit your prompt and mention what needs to be changed. For instance, if the character’s face doesn’t fit your requirement, mention “add realistic face anatomy, smooth skin, and avoid unnatural expressions” for improved results.

Do’s and Don’ts of Prompting

Do’s:

• Be specific since descriptive and clear prompts produce more realistic and high-quality results.
• Use visual styles like photorealism, cinematic, or editorial-style to acquire the desired tone and look in the generated images.
• Use camera vocabulary like "from the top" or "shallow depth of field" to enhance realism in your images.
• Use negative prompts to refine your images and exclude the unnecessary visual elements
• Reiterate and improve your prompt to get the desired images.

Don’ts:

• Don’t use ambiguous terms such as “beautiful,” “pretty,” or “cool” without elaborating the context.
• Don’t use contradicting phrases in the same prompt, as this might confuse the AI image generator. For instance, creating an image of “bright day with dark mood.”
• Don’t add multiple, conflicting styles in a singular prompt, as it might result in chaotic images.
• Don’t be discouraged if the results aren’t perfect on the first try. Reiterate, refine, and the model will create a realistic image you wish for.

Conclusion

Imagine 1.0 offers versatility and incredible accuracy when it comes to generating realistic visuals and even conceptual art. Unlock the full potential of this AI image generator by understanding prompt engineering and producing high-quality results.

Looking for more prompt guides? Read Runway Gen-4 Prompt Guide | Runway Act II Prompt Guide | Kling 2.1 Prompt Guide | PixVerse v5 Prompt Guide