I built a side project that turns baby photos into dancing videos using AI
I just launched AI BABY DANCER, a small side project that turns baby photos into short dancing videos using AI motion models.
It started as a fun experiment to explore how modern AI motion control can bring static images to life. After sharing a few test videos with friends, the reaction was unexpectedly positive — lots of laughs, shares, and “you should make this public”.
So I kept it simple:
Upload a baby photo
Pick a dance style
Generate a short, shareable video
No editing skills, no timeline, just something playful and fun.
Right now it’s very early:
Built as a side project
Focused on Web
Mostly free and experimental
I’d love feedback from other builders here:
Does this feel like something people would actually use?
Any thoughts on features or positioning?
Similar products you’ve seen that I should study?
Happy to answer any questions. Thanks for taking a look 🙏
This is such a fun concept I can totally see why it has viral potential. Curious how you balanced novelty vs repeatability here. Do users come back mainly to generate new dances, or is the first “wow” moment doing most of the work?
Also interested in how you thought about distribution early on was this built with social sharing as the core loop from day one, or did that come after launch?
Nice execution overall.
seems quite interesting
This idea might seem goofy, but it is actually well-crafted for sharing, thanks to the ingenious people behind it. ~
I’ve seen friends spend too much time in CapCut/Reels trying to funny edit baby or pet photos for birthdays and family groups. The creativity stems from the tools used. The editor was completely removed by you.
This clean and simple loop involves uploading a video, selecting a dance and getting the video.
To me, the strongest feeling is not “an AI tool” but a moment generator.
Birthday messages.
WhatsApp groups for family.
Milestones of baby.
Sharing by Grandparents on Facebook.
This is a pretty different positioning than “AI motion from images,” and probably closer to the reason why people would actually use it.
I’d be curious about features.
Themes that are already set up (b-day, x-mas, first steps etc.).
Caption or music overlay with a single click.
A hassle-free download and share to social networks.
This appears less like a builder tool and more like something that could go quietly viral with normal users if the output is funny enough.
You could be working on something that’s more than a side project.
非常厉害
That's an awesome project, it will help a lot of content creators
This sounds like a blast! There is something inherently hilarious about AI motion control, and applying it to baby photos is such a smart, "shareable" hook. Keeping it simple with a 3-step process (Upload → Pick → Generate) is exactly what users want for a playful side project.
Here are some thoughts on your questions:
Usage: Absolutely. High "virality" potential. This is the kind of thing parents and grandparents would immediately post to Instagram Stories or family group chats.
Positioning: You might consider positioning it as a "Digital Greeting Card" or a way to liven up "Month-by-Month" update photos that many parents already take.
Similar Products: While general AI motion tools like Luma Dream Machine or Kling AI exist, your niche focus on babies makes it much more accessible for non-techy users. You could also check out the Baby Dance AI app for inspiration on current mobile features.
Thank you for sharing this.
Thanks a lot, really appreciate the thoughtful feedback 🙏
I love the “digital greeting card” angle — that’s actually a use case I hadn’t explicitly framed yet, but it makes a ton of sense, especially for birthdays and monthly baby updates.
Totally agree on the niche focus too. There are plenty of powerful general AI motion tools, but most of them feel intimidating for non-technical users. My goal here is to make something parents can use without thinking about “AI” at all.
Thanks again for taking the time to write this — super helpful.
Congrats on the launch 👏
quick question — did you already set any hard limits on LLM/API spend?
I’ve seen a few indie AI apps get hit by agent loops early on.
Good call. We don’t run an autonomous LLM agent loop—each generation is a single user-triggered call to our video provider. We enforce server-side duration caps (10s image-led / 30s video-led) and require credits; job creation + credit deduction is atomic in the DB, so spend is bounded by purchased credits. Provider calls have bounded retries (max 3) and our status polling has a hard timeout (30min / 180 polls). Next up: adding a global daily budget / circuit breaker and rate limits for extra safety.
Congratulations on your launch. It looks impressive! What channels are you exploring to attract early users?
Thanks! Right now it’s very early, so I’m mostly experimenting.
The main focus so far has been organic sharing — people sending videos to friends, family group chats, or posting short clips on social platforms. I’m also curious to see how communities like Indie Hackers react before investing too much into paid channels.
Still very much in “learning mode” on this, so happy to hear any suggestions as well.
That makes a lot of sense, especially at this early stage. Organic sharing tends to be the best indicator of what truly resonates with people.
One approach that often works well in conjunction with this is utilizing Reddit communities as a learning resource, not for promotional purposes, but to test messaging related to the problem, use cases, or before-and-after stories. The feedback received can be surprisingly straightforward.
If you're interested, I can share my usual approach to using Reddit for early-stage tools to see what ideas stick before investing in ads.
Sounds interesting — I’d love to hear your approach. Thanks!
That's great to hear! I'm glad you find it useful.
To summarize, my approach is quite lightweight at this stage. I primarily use Reddit as a channel for feedback and validation. This usually involves testing discussion-style posts about the problem, various use cases, or before-and-after scenarios in relevant communities. I observe what topics people engage with the most before moving on to any promotional efforts. This method helps founders understand which messaging truly resonates from the outset.
If you're interested, we can discuss this in more detail, and I can customize the ideas to fit your product specifically. Feel free to reach out to me on WhatsApp at +1 365 782 8888 so we can communicate more easily.