Discover how to customize your starting image using Scene Builder to create the scene you want for your video.

Scene Builder is an additional tool in ActiveCast that helps you enhance your source image and add new elements to create the perfect image for your generated ActiveCast video.

Note. Each Scene generation requires additional credits. The number of credits deducted depends on the AI model you select. Nano Banana 2 Lite uses 45 credits per generation, while all other models use 90 credits per generation.

You’ll be redirected to the editor whenever you click Activate Image on a stock, uploaded, generated, or previously used image, or click Recreate on a showcase video.

Once you are in the editor, click “Add Element” above the preview screen and select one of the 4 available options:

Another way to access Scene Builder is to hover over the preview screen and click “Edit Image” button in the upper right corner of the screen.

Selecting “Create Scene” or “Edit Image” option opens Scene Builder tool directly.

If you select “Add Product”, “Add Location”, or “Add Character”, Scene Builder will open together with the corresponding library, where you can select an image from our library or upload your own. To close the library and return to Scene Builder, click “Back” in the upper-right corner.

You can add:

Character — Your main source (starting) image. This can be an uploaded photo or an image from our library.

Product — A photo of your product or merchandise from your uploaded images or similar photos from our library.

Click "Product" tab in the Scene Builder, find the image you like, hover over it and click "Add".

Location — A photo of the desired location. Scene Builder uses the background from this image when creating the final scene.

Click "Location" tab in the Scene Builder, find the image you like, hover over it and click "Add".

Reference Image — A photo of another person, object, or visual element that you want to incorporate into the final image.

Click "Reference Image" tab in the Scene Builder, find the image you like, hover over it and click "Add".

You can easily change the category assigned to an image at any time or replace each of those images. For example, you can change a Character image to a Product image or a Location image to a Product image. To do this, click the three dots in the upper-right corner of the image and select the desired category.

A Reference Image can only be replaced or deleted. Its category cannot be changed to Character, Product, or Location.

Choose from 3 available ratios:

Note. Not all AI models used for video generation support all image ratios. For example, Gemini Omni Flash, Veo, and LTX do not support the Square ratio.

For more information, see our article about the available AI models.

Use this field to describe the scene you want to create.

For the best results, provide an accurate and detailed description and include all the elements you want to appear in the final image. At the same time, avoid contradictory instructions, as they may confuse the AI and lead to unexpected results.

You can enter up to 3,500 characters in this field.

Pro Tip: You can use an AI assistant, such as ChatGPT, Gemini, or another tool, to help you create a prompt that reflects your request and desired result.

Here is the prompt we have used to generate this Scene. Note, this prompt was created by ChatGPT and slightly adjusted manually.

The ballet dancer dancing in the garden while holding the exact bag from the product photo in her hand. The man from the reference image is dancing near the ballet dancer. Do not add people from Location image. Match the lighting, color palette, mood, and joyful atmosphere of the reference image. Ensure the dancer, bag, man, and garden blend naturally without people in the location image, with realistic proportions, consistent lighting, and a refined, elegant composition. Preserve the bag’s original design and details in ballet dancer's hand.

Scene Style defines the overall visual aesthetic of the generated image and determines whether the result looks natural, polished, raw, or cinematic.

Lighting controls the atmosphere of the scene, including the overall brightness, contrast, warmth, and shadows.

If you want to match the Style and Lighting to your Reference Image simply choose "Match Reference" for both Scene Style and Lighting.

Level of Aesthetic Finish determines how refined the final image appears, ranging from a natural, authentic “Candid” look to a highly polished “Studio Polish” look.

Scene Builder uses separate AI models to generate your final image. Depending on your needs you can choose from 5 available AI models:

Note: The exact results may vary depending on your source images, prompt, and combination of settings you choose.

Before moving on to video generation, Scene Builder generates several images for you to choose from. To get better results, enable the Strong Variations option by turning on the toggle.

Once you have selected all the desired settings and ready to generate final image, click “Create Scene”. Scene Builder will generate several images that you can choose from.

Note. Each Scene generation requires additional credits. The number of credits deducted depends on the AI model you select. Nano Banana 2 Lite uses 45 credits per generation, while all other models use 90 credits per generation.

Select the generated image you like most and click “Continue to Video” to return to the editor.

If you are happy with your edited starting image, continue to the next article to learn how to set up your video generation settings, starting with Action Template and Camera Motion.

If you have any questions, feel free to contact us via in-app chat or email at [email protected].

We look forward to seeing your brand thrive with ActiveCast videos!