Skip to content
  • There are no suggestions because the search field is empty.

How do I use AI to identify assets from uploaded photos?

Nines can automatically identify Assets from one or more photos using AI. Simply upload photos of your Assets, review the suggested details, make any edits if needed, and add them to your Asset library.

1. Start a new AI Asset identification

From the Assets page, click Add new and select Identify from image.

Screenshot 2026-07-21 at 6.10.43 PM

2. Select a Property and upload photos

Choose the Property where the Assets belong, then upload your images by either:

  • Dragging and dropping them into the upload area
  • Clicking browse to select files from your computer

You can upload:

  • Up to 5 images at a time
  • JPEG, PNG, or WebP files
  • Maximum 5 MB per image

Click Identify assets to begin.

Screenshot 2026-07-21 at 6.10.37 PM

3. Review the identified Assets

After processing, Nines creates a draft for each identified Asset.

For each Asset, AI may suggest details such as:

  • Category
  • Asset name
  • Brand
  • Condition
  • Dimensions
  • Other identifying information, when available

Review each draft and make any necessary changes before adding it to your Asset inventory.

  • Click the + button to add an Asset.
  • Click the trash icon to discard an Asset you don't want to keep.

Screenshot 2026-07-21 at 6.15.17 PM

Tips for the best results

  • Use clear, well-lit photos.
  • Photograph one primary Asset per image whenever possible.
  • Include the entire Asset in the frame.
  • Avoid cluttered backgrounds or heavily obstructed objects.
  • If available, include close-up photos of labels or manufacturer badges to help improve identification.

Notes

  • AI identification provides a starting point and may not always be 100% accurate.
  • Always review the suggested details before adding an Asset.
  • If an Asset cannot be confidently identified, you can manually edit the information before saving it.

Security and Privacy

When a photo is sent to the AI model, the model analyzes the image using knowledge baked into its weights during training — it does not reach out to the internet, search external databases, or call any APIs at inference time. It’s pattern-matching against what it learned during training (product photos, catalogs, brand imagery, etc.).

No external lookup is happening. The model never “leaves” to go find information. Everything it knows about what a Hermès bag or a Viking range looks like was learned before deployment.

How to explain this simply:

“The AI was trained on a vast library of product imagery and catalog data. When it sees your photo, it recognizes the asset the same way an experienced estate manager would from memor, it’s not searching the internet or any outside database. Your photo is processed and then discarded; it’s not stored or shared.”