How do you use Lumion Cloud Ai Enhancer?
Enhance architectural images, compare results, and manage your creations in Lumion Cloud.
Lumion Cloud AI Enhancer uses generative AI to refine architectural visualizations directly within the Lumion Cloud workflow. Upload your render, select an AI model, then use either a Guided Mode or a custom Expert Mode prompt to enhance materials, lighting, atmosphere, and other visual details.
The AI Enhancer is especially useful for applying final polish, exploring alternative visual directions, or making targeted changes that would otherwise require returning to the original Scene or using separate image editing tools. Depending on the image and selected model, this could include refining materials, adding or removing visual elements, or correcting unwanted details and visual limitations that became apparent during the modeling or rendering process, without having to render the Scene again in Lumion Pro.
Enhanced results can be compared with the original, downloaded, or saved back to a Project as a new version or variation, keeping presentation alternatives organized within the same review and collaboration workflow.
Ready to try it yourself? Open Lumion Cloud and start enhancing your own architectural visualizations.
Menu:
|
1. Start Using the AI Enhancer |
How the AI Enhancer works
The AI Enhancer processes your image using one of two AI models:
Both AI models support two workflows:
-
Guided Mode - provides predefined controls for preserving elements and adjusting the time of day.
-
Expert Mode - lets you describe the required enhancement using a custom AI prompt.
The scene below uses Expert Mode to demonstrate how the two AI models can interpret the same custom prompt differently.
The same source image (below) and Japanese-themed prompt were used with each AI model. Although the AI Enhancer is designed to preserve the original composition and architectural intent, generative AI may still alter geometry or other details.
Example Scene: Farnsworth

You can start by uploading a local image directly to the AI Enhancer, or by opening an image that is already stored in Lumion Cloud.
1.1 Select an image via Browse or Drag and Drop
Open the AI Enhancer from the Lumion Cloud navigation toolbar. Depending on the toolbar view, select either the icon or the expanded AI Enhancer menu option.
Supported files: The AI Enhancer processes still images only. You can upload a JPG, JPEG, or PNG file up to 50 MB. Other file types are not supported.
(Expanded Toolbar View)
From the AI Enhancer page, drag and drop an image into the upload area or select Browse local files.
The upload page also displays your recent AI Enhancer creations. These results cannot be enhanced again directly from your Creations. To reuse one, first save a copy to a Project or Renders folder.
1.2 Select an image already in your Project or Renders folder
Alternatively, open an existing image from the Renders folder or from a Project. From the image preview, you can access the available image-refinement tools:
Important: Starting a new AI Edit overwrites any current edits applied to the image. Save a copy before proceeding if you want to retain them. AI Edit actions cannot currently be stacked within the same window process.
1.2.1 Select AI Edits
Select the AI Edits button to open the available AI editing tools.

1.2.2: Choose Enhance or Upscale:
The AI Edits window provides two options:
-
Enhance - uses generative AI to refine lighting, color, detail, and other visual elements.
-
Upscale - increases image resolution while maintaining sharpness.

For more information about the Lumion Cloud AI Upscaler, check out this article:
-
Knowledge Base: How do you use the AI Upscaler in Lumion Cloud?
To continue with the AI Enhancer, select Enhance, then select Start AI Enhancement. This opens the AI Enhancer, where you can select an AI model, check the image compatibility, and configure the enhancement settings.
1.2.3 Post-Processing editor 
The image preview also includes the new Post-Processing editor for non-generative adjustments such as cropping, straightening, image adjustments, presets, and LUTs.
The Post-Processing editor is separate from the AI Edits workflow and is not required to continue with the AI Enhancer. To find out more, see this article:
-
Knowledge Base: How do you use Post-Processing in Lumion Cloud?
After selecting an AI model, Lumion Cloud checks the uploaded image against that model’s processing requirements. The available checks and limits vary by model.
2.1 Select an AI model
FLUX [Pro] |
ChatGPT Images 2 |

2.2 Compatibility status
Each requirement is highlighted according to its current status.
![]() |
|
2.3 Adjusting an image with Fix It (If required)
Most images will either meet the selected model’s requirements or be blocked for falling outside its supported size limits.
In some instances, Lumion Cloud displays a Fix it button if your image dimensions are not divisible by 16. This option appears only when the image can be adjusted to compatible dimensions using the built-in crop tool.
Important: The crop controls are only available after Fix it has been selected. Review the crop before selecting Continue, as only the area inside the crop frame will be sent for processing.
2.3.1 How each model handles the warning
The way this compatibility issue is handled depends on the selected AI model:
-
With FLUX [Pro], the warning is highlighted in orange. You can use Fix it button, or continue without manually adjusting the image. The FLUX model can accept a slight difference from the preferred dimensions, although the output is not guaranteed to match the original dimensions.

-
With ChatGPT Images 2, the issue is highlighted in red because the model has stricter dimension requirements. You must select Fix it and confirm a compatible crop before continuing.

2.3.2 Adjusting the crop
After selecting Fix it, the crop tool opens and suggests a balanced crop using dimensions supported by the selected AI model.
Before continuing, you can:
-
Move the crop frame to select a different part of the image.
-
Move the image behind the crop frame.
-
Resize the crop frame horizontally, vertically, or diagonally.
-
Review the resulting pixel dimensions as you make adjustments at the bottom of the crop window.
The crop dimensions snap to values supported by the selected AI model (Divisible by 16 values). This allows you to control which parts of the image are retained and preserve the original composition and aspect ratio as closely as possible.
For example, a 1920 × 1080 image may be cropped to 1920 × 1072 by removing eight pixels evenly on the vertical axis. Alternatively, you could reduce both dimensions to another compatible size, such as 1904 × 1072, to retain proportions closer to the original 16:9 aspect ratio.
2.3.3 Reviewing and confirming the crop
Use the controls in the crop window to review the image before continuing:
- Zoom controls: Inspect the image and crop more closely.
- Reset view: Return the preview to its default position and zoom level.
- Start over: Discard the current adjustments and return to the initially suggested crop.
- Continue: Confirm the crop and proceed to the enhancement settings. After selecting Continue, the selected AI model is locked and cannot currently be changed for that enhancement.
Important: Once you select Continue and pass the compatibility stage, the selected AI model becomes locked for that enhancement and cannot currently be changed.
After confirming image compatibility, choose how to enhance the image. You can use Guided Mode settings or enter your own instructions in Expert Mode.
3.1 Guided Mode:
This mode provides predefined controls for preserving selected elements and adjusting the time of day. Element preservation helps protect important parts of the original design while allowing other areas of the image to be enhanced.
|
|
Elements to Preserve
Adjust Time of Day
|
This makes it possible to explore different presentation moods without having to return to Lumion Pro to produce another render.

3.2 Expert Mode:
Expert Mode lets you describe the intended result using your own instructions. You can request subtle refinements to lighting and materials, broader changes to the atmosphere or landscaping, or changes to specific visual elements.
Some ideas to consider when writing your prompt include:
- Materials, textures, and surface details: Refine concrete, timber, metal, glass, fabric, paving, or other surfaces. You can also request more realistic reflections, weathering, texture variation, or finer material detail.
- Lighting, time of day, weather, and atmosphere: Change the scene to sunrise, dusk, night, or another lighting condition. Consider interior illumination, shadows, cloud cover, fog, rain, or a warmer or cooler atmosphere.
- Vegetation, furniture, people, and other scene elements: Add, remove, or refine landscaping, outdoor furniture, vehicles, decorative objects, or people. You can also describe their placement, scale, density, or activity.
- Color balance, realism, and overall visual style: Adjust contrast, warmth, saturation, mood, or photographic realism. You can also describe a specific presentation style, such as cinematic, minimalist, natural, dramatic, or editorial.
- Elements that should remain unchanged: Identify important features such as the building geometry, camera position, perspective, roofline, openings, materials, or access routes that should be preserved.
|
|
Custom AI Prompt
|
Disclaimer: References to copyrighted material in the prompt may be moderated and rejected by the model.
Tips for better prompts
The 6,000-character limit is a maximum. Clear, specific instructions generally work better than a long list of loosely connected changes.
A useful prompt structure is:
"Preserve [important elements]. Enhance [materials, lighting, or details]. Add or refine [desired elements]. Create [atmosphere or style]. Avoid changing [restrictions]."
For example:
"Preserve the building geometry, camera position, roofline, windows, and doors. Refine the concrete, timber, and glazing materials, and create warm dusk lighting with subtle interior illumination. Add restrained landscaping around the entrance. Maintain realistic scale and straight architectural lines. Do not add people or alter the building openings."
Select
to begin processing the image. Once processing begins, you must wait for the enhancement to complete.

Note: Processing times can vary between models.
After selecting Enhance, Lumion Cloud begins processing the image. Once complete, you can review the enhanced result, compare it with the original, generate another variation, or save and download the image.
4.1 Review and compare the result
Once processing is complete, the enhanced result is displayed alongside the original image.

Use the Original and Enhanced thumbnails on the left to switch between the two versions. You can also enable Compare to:
-
Drag the divider across the image to inspect the differences.
-
Switch between horizontal and vertical comparison.
-
Display only the Original or Enhanced image
-
Zoom in or out.
-
Reset the preview.
-
Open the image in full screen.
4.2 Save or continue
After reviewing the result, you can:
-
Select Generate another, to create a new variation.
-
Select Go to My Creations, to open your AI Enhancer gallery.
-
Select Save copy to, to save the result to a Project or Renders folder.
-
Select Download, to download the selected image.
-
Select Rate, to provide feedback on the result.
-
Open the More (3 dots) menu to save a copy, rename, or delete the creation.
Recent AI Enhancer results appear under Your Creations. Select View all, or go to Your Creations, to open the complete gallery where you can browse, sort, and manage your enhanced results.
5.1 Browse and sort your Creations
The gallery displays each enhanced result as an individual tile. You can sort the gallery by:

-
Last modified
-
Created date
-
Name (A-Z)
-
Name (Z-A)
If you need to create another enhancement, select New generate.
5.2 Manage an individual creation
From the gallery, open the three-dot menu on any creation tile to:

-
Select Save copy to, to save the image to a Project or Renders folder.
-
Select Download to save the image locally.
-
Select Rename to change the creation name.
-
Select Delete to remove it from your Creations.
You do not need to open the creation before using these options.
5.3 Open and review a creation
Select a creation tile to open it in the detailed view. The carousel at the top provides quick access to your other creations without returning to the gallery.

From the detailed view, you can:
- Select another creation from the carousel.
- Switch between the Original and Enhanced images.
- Use Compare to inspect the differences between each image.
- Select Save copy to, to save the selected creation to a Project or Renders folder.
- Select Download to save the selected image locally.
To enhance an existing creation again, first select Save copy to and save it to a Project or Renders folder. Open the saved image, select AI Edits, and then choose Enhance.
For more information on browsing and managing Projects in Lumion Cloud, see:
-
Knowledge Base: Lumion Cloud: Projects Screen

6. Current limitations
-
This is built specifically for architectural visualization. Close-up subjects without architectural context (like personal portraits or pet photos) won't work well.
-
Results and time to render may vary between models and generations.
-
Generative AI may change details unexpectedly.
-
Maximum output resolution is 4K.
-
It will retain the original resolution rather than upscale every image.
-
The selected model cannot be changed after compatibility is completed.
-
Settings cannot currently be reopened after generating the result.
7. See Also
-
Knowledge Base: How do you use Post-Processing in Lumion Cloud?
-
Knowledge Base: Lumion Cloud: Security and Privacy FAQ
-
Knowledge Base: How do you use the AI Upscaler in Lumion Cloud?
-
Knowledge Base: How do you use the Lumion Cloud AI Material Generator
-
Knowledge Base: What is Lumion Cloud?
-
Knowledge Base: Lumion Cloud: Render Screen
-
Knowledge Base: Lumion Cloud Video Tutorial 1: Lumion Cloud Overview



