Reference

Image-to-prompt generator workflow for Midjourney prompts from images

This image-to-prompt generator example is built around a Midjourney prompt from image workflow: RefPrompt extracts the reference's architecture, lighting, material mood, and composition, then reshapes it into Midjourney-friendly language plus reusable video and negative prompts.

RefPrompt treats the image as a visual grammar source, then compresses that grammar into the concise nouns, materials, lighting cues, and aspect-ratio controls Midjourney responds to best.

Midjourney-firstImage promptArchitectureStyle extraction

What this workflow covers

Follow the reference analysis through reusable image and video prompts, a Runway adaptation, a real result, practical limitations, and focused negative constraints.

Minimal architectural interior used for a Midjourney prompt from image workflow — Reference direction: quiet architecture, muted materials, strong planes, and gallery-like stillness.

Use your own reference

Turn your own visual into a structured prompt.

Start with an image in the free web tool, or use the Chrome extension on an image or video already open in your browser.

Open the free RefPrompt tool Use the Chrome extension

Style Breakdown

How the reference is structured

The style breakdown focuses on reusable design logic: geometry, negative space, material restraint, and light direction.

Composition: Large architectural planes, centered human-scale calm, and enough negative space to make the scene feel premium instead of busy.
Materials: Concrete, warm wood, soft plaster, and matte surfaces create tactile contrast without glossy product-ad shine.
Light: Diffuse natural light enters from the side, producing soft shadow blocks and a slow editorial rhythm.
Promptable style: The reference works best as a minimal editorial interior, not as a hyper-detailed fantasy environment.

Image Prompt

Image Prompt for Midjourney prompt from image output

The image prompt expands the reference into complete still-generation language.

Minimal architectural interior with sculptural concrete planes, warm wood accents, soft plaster walls, diffuse side light, quiet gallery atmosphere, restrained neutral palette, human-scale calm, editorial composition, clean negative space, tactile matte materials, refined architectural photography.

Video Prompt

The video version keeps the same environment but adds a slow reveal and continuity.

Slow interior reveal through a minimal architectural space, camera gliding from a shadowed foreground plane into a softly lit room. Maintain matte concrete and wood textures, side-lit shadows, quiet gallery atmosphere, and stable geometry. Motion should feel like an architectural film: smooth, slow, and observational, with no sudden cuts or decorative effects.

Midjourney Version

This is the shortest and most visual form of the reference, tuned for Midjourney scanning.

minimal architectural interior, sculptural concrete planes, warm wood accents, soft plaster, diffuse side light, quiet gallery atmosphere, restrained neutrals, editorial architecture photography, clean negative space --ar 4:5 --style raw --v 6

Runway Version

Runway receives the same reference with camera movement, reveal timing, and stable spatial constraints.

Use the image as a reference for a slow 5-second architectural reveal. Begin close to a dark foreground plane, then glide right to reveal the warm wood and softly lit concrete space. Keep the room geometry consistent, preserve matte textures, maintain natural side light, and avoid adding people, furniture clutter, lens warping, or dramatic lighting changes.

Result & Limits

Midjourney prompt from image result for a minimal interior

The image-to-prompt generator should not describe every object in the room. The useful extraction is the visual grammar: large planes, matte concrete, warm wood, diffuse side light, restrained palette, clean negative space, and quiet architectural stillness.

Generated minimal architectural interior result from a Midjourney prompt from image case study — Generated result: a restrained architectural interior with concrete, warm wood, soft plaster, and clean negative space.

Prompt used in this case

The Midjourney version below compresses the longer image-to-prompt analysis into compact visual language: minimal architectural interior, sculptural concrete planes, warm wood, soft plaster, side light, quiet gallery atmosphere, restrained neutrals, and --style raw.

Result read

The generated result keeps the reference's calm architecture, material contrast, and side-lit planes. It is useful as a Midjourney exploration frame, but the prompt still needs negative constraints for perspective, clutter, and fake glossy materials.

What worked

The short Midjourney prompt preserved material contrast without overexplaining the scene.
Side light and strong planes survived better than small decorative details.
The result demonstrates why image-to-prompt extraction should focus on reusable visual decisions, not literal object lists.

Errors and limitations

Perspective control: Architectural prompts expose crooked walls quickly. Add perspective and geometry negatives when the output must feel real.
Material drift: Concrete, plaster, and wood can become glossy or synthetic unless the prompt names matte, tactile surfaces.
Reference compression: Midjourney often works better with concise cues, but over-compressing can lose the reference's spatial rhythm.

Negative Prompt

For Midjourney and video tools, the main risks are clutter, fake materials, and unstable geometry.

cluttered furniture, glossy plastic surfaces, distorted walls, crooked perspective, oversaturated colors, fake HDR, harsh spotlight, busy decor, messy cables, low-resolution concrete texture, warped door frames, fisheye lens, extra windows, cartoon rendering, noisy shadows.

CTA

Reference in, structured prompts out.

Use the same page structure inside RefPrompt: start from a visual reference, then copy the prompt version that fits the tool and channel.

Try this workflow with RefPrompt.

Questions about this workflow

How do I create a Midjourney prompt from image references?

Start by extracting the reference's subject, materials, lighting, composition, mood, and camera language. Then compress those details into Midjourney-friendly phrasing with aspect ratio and style controls, as shown in the Midjourney Version section.

Why is the Midjourney prompt shorter than the full image prompt?

Midjourney responds well to compact visual cues. The longer image prompt is useful for general tools, while the Midjourney version removes extra explanation and keeps the strongest nouns, materials, light, and composition signals.

Can this Midjourney prompt from image workflow be reused for video tools?

Yes. The same reference can become a Runway or Kling prompt by adding camera movement, pacing, scene continuity, and negative constraints that protect geometry and material consistency.