Use Gemini Omni for a home-design workflow

Step by step

Run the workflow

1. 01

   Plan the transformation

   Use the source inputs to complete this stage and produce a result that can be checked before continuing.

   1. Your AI output is only as strong as your original source.
   2. Conceptualize how you want your final video to look.
   3. source footage that acts out the scenes you want to create. 2.You can replace
   4. the environment, actor, or visual style later using AI.

   

   Common failure modes

   • The model changes the subject, framing, text, product, or background instead of editing only the requested element.
   • The result looks polished at first glance but contains broken hands, unreadable text, warped objects, or inconsistent lighting.
2. 02

   Generate the image first!

   Use the source inputs to complete this stage and produce a result that can be checked before continuing.

   • Click here to open Google Flow (opens in a new tab)Direct link supplied in the source. Source page 4.

   1. Generate the image first! With your assets ready, it's time to edit your footages!
   2. Go to Google Flow and select "Create with Flow."
   3. Select + New project

   

   

   

   Common failure modes

   • The model changes the subject, framing, text, product, or background instead of editing only the requested element.
   • The result looks polished at first glance but contains broken hands, unreadable text, warped objects, or inconsistent lighting.
3. 03

   Upload the source asset 1

   Use the source inputs to complete this stage and produce a result that can be checked before continuing.

   1. You may receive free credits (varies by account).
   2. Upload your source asset.
   3. After you uploaded your asset, now fill in the prompts.

   

   

   

   Common failure modes

   • The model changes the subject, framing, text, product, or background instead of editing only the requested element.
   • The result looks polished at first glance but contains broken hands, unreadable text, warped objects, or inconsistent lighting.
4. 04

   Upload the source asset 2

   Use the source inputs to complete this stage and produce a result that can be checked before continuing.

   1. Copy the prompt below and replace any project-specific subject, product, location, or format details.
   2. Attach the source assets in the same order described by the prompt, then run one controlled test.

   Working asset

   Generate an image that changes the home design to a Japandi-style interior, but do not change the person in the image; you can change the chair design he is sitting on. You can change the ceiling fan and lights, as well as the furniture

   

   

   Common failure modes

   • The model changes the subject, framing, text, product, or background instead of editing only the requested element.
   • The result looks polished at first glance but contains broken hands, unreadable text, warped objects, or inconsistent lighting.
5. 05

   Upload the source asset 3

   Use the source inputs to complete this stage and produce a result that can be checked before continuing.

   1. Copy the prompt below and replace any project-specific subject, product, location, or format details.
   2. Attach the source assets in the same order described by the prompt, then run one controlled test.

   Working asset

   Generate an image that changes the home design to a Japandi-style interior, but do not change the person in the image; you can change the chair design he is sitting on. The man is holding on a cup of tea

   

   

   Common failure modes

   • The model changes the subject, framing, text, product, or background instead of editing only the requested element.
   • The result looks polished at first glance but contains broken hands, unreadable text, warped objects, or inconsistent lighting.
6. 06

   Upload the source asset 4

   Use the source inputs to complete this stage and produce a result that can be checked before continuing.

   1. Generate an image that changes the kitchen design to a Japandi-style interior, but do not change the person in the image

   

   

   Common failure modes

   • The model changes the subject, framing, text, product, or background instead of editing only the requested element.
   • The result looks polished at first glance but contains broken hands, unreadable text, warped objects, or inconsistent lighting.
7. 07

   Upload the source asset 5

   Use the source inputs to complete this stage and produce a result that can be checked before continuing.

   1. Copy the prompt below and replace any project-specific subject, product, location, or format details.
   2. Attach the source assets in the same order described by the prompt, then run one controlled test.

   Working asset

   Generate an image that changes the toilet design to a Japandi-style interior, but do not change the person in the image. You change the toilet bowl design he sat on and the hangers. Also, clean up the toilet to make it look neater, and remove all the hanging buckets.

   

   

   Common failure modes

   • The model changes the subject, framing, text, product, or background instead of editing only the requested element.
   • The result looks polished at first glance but contains broken hands, unreadable text, warped objects, or inconsistent lighting.
8. 08

   Upload the source asset 6

   Use the source inputs to complete this stage and produce a result that can be checked before continuing.

   1. Copy the prompt below and replace any project-specific subject, product, location, or format details.
   2. Attach the source assets in the same order described by the prompt, then run one controlled test.

   Working asset

   Generate an image that changes the bedroom design to a Japandi-style interior, but do not change the person in the image. Change the bed design he sat on, the ceiling fan, the ceiling, and the window grill. Make the house neater.

   

   

   Common failure modes

   • The model changes the subject, framing, text, product, or background instead of editing only the requested element.
   • The result looks polished at first glance but contains broken hands, unreadable text, warped objects, or inconsistent lighting.
9. 09

   Assemble the video

   Use the source inputs to complete this stage and produce a result that can be checked before continuing.

   1. Now Edit your video with your newly generated image.
   2. With your assets ready, it's time to edit your footages!
   3. Original Video AI Generated Image

   

   

   

   Common failure modes

   • The model changes the subject, framing, text, product, or background instead of editing only the requested element.
   • The result looks polished at first glance but contains broken hands, unreadable text, warped objects, or inconsistent lighting.
10. 10

    Generate a transformed scene 1

    Use the source inputs to complete this stage and produce a result that can be checked before continuing.

    1. Original Video AI Generated Image
    2. Transform the house into the reference photo and follow the motion of the man in the video

    

    

    Common failure modes

    • The model changes the subject, framing, text, product, or background instead of editing only the requested element.
    • The result looks polished at first glance but contains broken hands, unreadable text, warped objects, or inconsistent lighting.
11. 11

    Generate a transformed scene 2

    Use the source inputs to complete this stage and produce a result that can be checked before continuing.

    1. Original Video AI Generated Image
    2. Transform the house into the reference photo and follow the motion of the man exactly in the video. Do not change how the man looks.

    

    

    Common failure modes

    • The model changes the subject, framing, text, product, or background instead of editing only the requested element.
    • The result looks polished at first glance but contains broken hands, unreadable text, warped objects, or inconsistent lighting.
12. 12

    Generate a transformed scene 3

    Use the source inputs to complete this stage and produce a result that can be checked before continuing.

    1. Original Video AI Generated Image
    2. Transform the kitchen into the reference photo and follow the motion of the man exactly in the video. Do not change how the man looks.

    

    

    Common failure modes

    • The model changes the subject, framing, text, product, or background instead of editing only the requested element.
    • The result looks polished at first glance but contains broken hands, unreadable text, warped objects, or inconsistent lighting.
13. 13

    Generate a transformed scene 4

    Use the source inputs to complete this stage and produce a result that can be checked before continuing.

    1. Original Video AI Generated Image
    2. Transform the toilet into the reference photo and follow the motion of the man exactly in the video. Do not change how the man looks from the video

    

    

    Common failure modes

    • The model changes the subject, framing, text, product, or background instead of editing only the requested element.
    • The result looks polished at first glance but contains broken hands, unreadable text, warped objects, or inconsistent lighting.
14. 14

    Generate a transformed scene 5

    Use the source inputs to complete this stage and produce a result that can be checked before continuing.

    1. Original Video AI Generated Image
    2. Transform the bedroom into the reference photo and follow the motion of the man exactly in the video. Do not change how the man looks from the video

    

    

    Common failure modes

    • The model changes the subject, framing, text, product, or background instead of editing only the requested element.
    • The result looks polished at first glance but contains broken hands, unreadable text, warped objects, or inconsistent lighting.

Before you use it

Final checks

• Check the full result for changed people, products, text, timing, or layout.
• Confirm the tool, model, controls, and cost before generating.
• Keep private or unreleased material out of third-party tools.