Cubigo Nova can generate an image for an activity directly in the back office, so you don't have to search a stock library or upload your own photo every time you create an activity.
Table of contents
- What this feature does
- Availability
- How to use it
- What Nova will and won't put in the image
- Tips for better results
- Frequently asked questions
- Getting help
1. What this feature does
Nova reads the details you've already entered for the activity and turns them into a realistic image you can attach to the activity. If you don't like the result, you can adjust the wording and generate again.
This is separate from AI image suggestions, an earlier feature that pulls keywords from your activity and returns matching photos from external royalty-free libraries. Image suggestions find an existing photo. Image generation creates a new one. Some communities have both.
You can still upload your own photo or use your community's image library at any time. Nova is an option, not a replacement.
2. Availability
Activity image generation is enabled per community and is not on by default.
If you'd like it turned on for your community, contact your Customer Experience Manager or raise a request through Cubigo Support.
If your organization has an internal review process for AI features, that review should happen before activation. Cubigo will not enable AI features at a community that has asked us to hold them.
3. How to use it
- In the back office, go to Activities and open the activity you want an image for, or create a new one.
- Fill in the activity title and description. Nova works from these, so a good activity description leads to a good generated image.
- Once that information is filled out, the Generate image button appears in color. Select it.
- Nova writes an image description. This is the prompt Nova will use to create the image, not the image itself.
- Review that image description and edit it if it isn't what you want. This is your main control over the result, so it's worth reading before you continue.
- Generate the image. It typically takes around 30 seconds to completely generate.
- If the image isn't right, modify the image description and regenerate. You can repeat this as often as you need to curate the image toward what you're looking for.
- Save the activity once you're happy with the image.
4. What Nova will and won't put in the image
- People are excluded. Nova is instructed not to render faces or bodies, including blurred people in the background.
- No literal time or place. Nova won't try to depict your specific room, building, or time of day.
- Realistic style. Nova aims for a warm, inviting, professional photography look. There is currently no style selector.
5. Tips for better results
- Describe the subject, not the event. "A bingo card and wooden markers on a table" gives a cleaner result than "residents playing bingo."
- Be concrete about objects. Named objects render far better than moods or abstractions.
- Keep it to one scene. Asking for several things at once tends to produce cluttered images.
- Iterate in small steps. Change one element per attempt so you can tell what worked.
- Expect a few tries for anything complex. Simple, object-focused activities generate well on the first attempt. Detailed or specific scenes take fine-tuning.
6. Frequently asked questions
Can I still upload my own images?
Yes. Nothing changes about uploading your own photos or using your community's image library.
Will residents know the image was AI-generated?
There's no automatic label on the image. If your organization has a disclosure policy for AI-generated content, apply it as you would elsewhere.
Is any resident data used to create the image?
No. Nova uses the activity's own details, meaning the title and description you enter. No resident information is used, and no residents appear in generated images.
Can I use this for dining menus?
Not today. Image generation for dining has been requested and is tracked as a product idea, but it isn't available.
Why does my image look nothing like our community?
Nova generates a generic representative image, not a photo of your actual space. If you need your real rooms and residents, please upload your own photos.
I got an error message instead of an image. What happened?
The message reads: "Unable to generate image. The provided description is too vague, unsafe, or violates content guidelines. Please revise and try again."
The two common causes are asking for people and writing a description that is too short. Nova will not generate faces or bodies, so any request for residents, staff, or a crowd returns this error. If your activity needs people in the shot, upload your own photo instead. If you weren't asking for people, add more concrete detail about the objects and setting and try again.
7. Getting help
Contact your Customer Experience Manager or submit a ticket to Cubigo Support with the activity name, the description you used, and a screenshot of the result.
Comments
0 comments
Article is closed for comments.