I'm Ding Yu, a product manager/software engineer from Tokyo, and I'd like to share a side project I built: Shots Maker (https://shotsmaker.com/). It's an app that lets you upload a photo of your garment, pick a model, and instantly generate a realistic image of the model wearing your product.
Why I built this:
The idea came from a friend who's a fashion designer. They were frustrated with the time and cost involved in traditional photoshoots. So I started building this tool after work as part of a "12 startups in 12 months" challenge.
How it works:
1. Upload a photo of a garment (e.g., a t-shirt or dress).
2. Choose a fashion model that suits your style.
3. Get a realistic image of the model wearing your clothes, on a clean white background.
Challenges & Lessons:
1. It's not yet possible to 100% duplicate the exact textures and patterns, so only simpler clothes work well right now. I've spent a lot of time tackling this issue, even using OpenAI's API to pick the best image from multiple candidates.
2. Fashion photoshoots are complex! Some clients want dynamic poses, unique backgrounds, or other creative elements that go beyond what my app currently offers. I'm working on adding more templates, but right now, it's focused on e-commerce settings—simple, clean, and professional product images.
What's next:
1. Adding more model diversity and poses.
2. Exploring ways to incorporate custom backgrounds for more versatile looks.
I'm Ding Yu, a product manager/software engineer from Tokyo, and I'd like to share a side project I built: Shots Maker (https://shotsmaker.com/). It's an app that lets you upload a photo of your garment, pick a model, and instantly generate a realistic image of the model wearing your product.
Why I built this: The idea came from a friend who's a fashion designer. They were frustrated with the time and cost involved in traditional photoshoots. So I started building this tool after work as part of a "12 startups in 12 months" challenge.
How it works: 1. Upload a photo of a garment (e.g., a t-shirt or dress). 2. Choose a fashion model that suits your style. 3. Get a realistic image of the model wearing your clothes, on a clean white background.
Challenges & Lessons: 1. It's not yet possible to 100% duplicate the exact textures and patterns, so only simpler clothes work well right now. I've spent a lot of time tackling this issue, even using OpenAI's API to pick the best image from multiple candidates. 2. Fashion photoshoots are complex! Some clients want dynamic poses, unique backgrounds, or other creative elements that go beyond what my app currently offers. I'm working on adding more templates, but right now, it's focused on e-commerce settings—simple, clean, and professional product images.
What's next: 1. Adding more model diversity and poses. 2. Exploring ways to incorporate custom backgrounds for more versatile looks.
If you're curious: You can check out the website (https://shotsmaker.com/) or the demo video (https://www.youtube.com/watch?v=DMcie9yuuC8). Any feedback would be awesome, especially if you're in the e-commerce space or have thoughts on AI-generated images.
Thanks for reading.
:wq