Tutorial · Getting Started
Short answer: To make a Smalltown Boy AI video from two photos, choose clear images, upload them to the generator, check the available output settings, and preview the result before sharing. Follow the live interface if its steps differ from this general workflow.
Content updated: October 10, 2026
Two photos can be a fun starting point for a personalized Smalltown Boy AI video: imagine a clip featuring you and a friend, two family members, or two pets. The exact steps depend on the generator you choose, but good preparation makes the process easier and helps you avoid common upload mistakes.
1. Choose the photos before opening the generator
Pick images where each subject is easy to identify. Clear lighting, visible faces, and enough space around the subject are useful starting points. Avoid heavily blurred images, extreme close-ups that cut off important features, or photos with several people if you need the tool to distinguish specific individuals.
Think about how the two images will work together. Similar framing and lighting can make the pair feel more consistent, although the generator may still interpret each image differently. There is no photo setup that guarantees a perfect result.
2. Check the tool's current requirements
Visit the Smalltown Boy AI generator and read the current instructions before uploading anything. Confirm what inputs the workflow requests, which file types and sizes are supported, whether an account is needed, and whether generation uses credits. Do not assume that every AI video tool accepts two photos in the same way.
3. Follow the on-page workflow
- Open the generator and locate the relevant creation workflow.
- Upload the requested photos in the order indicated by the interface.
- Review any settings or instructions shown on the page.
- Check the displayed cost or credit use before starting generation.
- Wait for the result, then preview it before downloading or sharing.
These are general steps, not a promise that every interface uses identical labels. If the page requests different inputs, follow the live instructions rather than trying to force a two-photo workflow.
4. Review the result carefully
Look at the faces, body proportions, framing, and motion throughout the clip. AI-generated movement can introduce visual glitches, especially around hands, limbs, fur, or overlapping subjects. If the result is not usable, review the tool's troubleshooting guidance before spending more credits on another attempt.
The videos covered by this workflow are up to 15 seconds long, so plan one clear visual idea rather than trying to tell a complicated story in a single clip.
Common problems and simple adjustments
- A subject looks unclear: try a sharper image with better lighting and a more visible face.
- The framing feels awkward: use a photo with more space around the subject.
- The two subjects look inconsistent: choose images with reasonably similar lighting and visual scale.
- The upload fails: check the current file requirements and try a supported image format.
- The output has glitches: remember that generation is probabilistic; changing the input may help, but cannot guarantee a fix.
Respect privacy and permission
Only upload photos you have the right to use. If the image shows a friend, partner, colleague, or another private individual, get their permission before uploading it and before posting the result. Be especially careful with children's photos and any context that could embarrass or misrepresent someone.
Ready to try it?
Prepare your images first, then check the live workflow and its terms.
Product interfaces, supported inputs, pricing, and privacy terms can change. Verify them on the live site before publishing or relying on specific instructions.
A quick pre-generation checklist
Before you upload, open both images side by side and check that the intended subject is obvious in each. Remove accidental screenshots, distracting borders, and unnecessary background clutter if you can do so without changing the person's or animal's appearance. Keep an untouched copy of each original photo in case you need to try a different crop.
After generation, evaluate the clip in three passes: check the first and last frames for awkward cropping, watch the movement for visual glitches, then review the full clip at normal speed. If you need to retry, change one variable at a time—such as using a clearer photo—so you can tell what actually helped. The live product instructions take precedence if its workflow differs from this general checklist.
Frequently Asked Questions
Do the two photos need to look similar?
They do not need to be identical, but clear, well-lit images with visible subjects make it easier to judge whether the result matches your idea.
What if the generator’s steps differ from this guide?
Follow the live product interface and its current requirements. This article describes a general workflow, not a guarantee that every version uses the same controls.
Related guides
Continue with a guide that answers the next question in the process: