How many reference photos do I need?

Two photos — one for each person in the kiss-cam scene. Each photo should be a clear, front-facing portrait.

Will the faces look accurate?

Yes. The template uses reference-image-based generation that preserves facial features with high accuracy. Well-lit, front-facing photos produce the best identity match.

Can I customize the stadium or event setting?

Yes. The default prompt creates a classic stadium kiss-cam with a red heart frame. You can modify the prompt to change the venue, frame style, or crowd energy.

AI Video Template

Valentine Kiss-Cam

Recreate the classic stadium kiss-cam moment with AI. Upload two face photos and the template generates a realistic arena scene — heart frame, cheering crowd, and a sweet on-screen moment between the two subjects.

Use this template See how it works

2,182,705+ happy users

Why use this template

Skip the blank page — start from a structure that works.

A fun, personalized video gift without needing to be at a live event.

Realistic stadium lighting and crowd atmosphere sell the illusion.

Both subjects keep their original appearance with accurate face preservation.

How to get the best output

Follow these steps for production-ready results every time.

Upload two clear, front-facing reference photos — one for each person.

Review the default prompt and adjust the stadium setting if desired.

Generate the kiss-cam video and check face accuracy for both subjects.

Download and share directly or add to a greeting card or social post.

Try it now — it's free

Best for

Create personalized Valentine's Day or anniversary video gifts.
Produce shareable social media content around sports events and holidays.
Build engagement content for dating apps or relationship-focused brands.

Prompt tips

Use well-lit, forward-facing portraits — side profiles reduce face accuracy.
Photos with neutral expressions work best since the AI adds the smile and lean-in.
High-resolution reference images (at least 512×512) produce sharper face details.

Frequently asked questions

: Two photos — one for each person in the kiss-cam scene. Each photo should be a clear, front-facing portrait.
: Yes. The template uses reference-image-based generation that preserves facial features with high accuracy. Well-lit, front-facing photos produce the best identity match.
: Yes. The default prompt creates a classic stadium kiss-cam with a red heart frame. You can modify the prompt to change the venue, frame style, or crowd energy.