Heyo!
So, there's been a bunch of new features added, as written in the What's Next section.
All changes have been pushed to production, so please check 'em out at the app link.
Feedback is much appreciated!
A very concise changelog:
- Now, there's a custom Express backend, with complete user authentication, a storage vendor, mail verification & a cloud managed database all connected within.
- I found out that running the face detection model in-browser on mobile devices could be very tasking and end up freezing all processes. To that end, surely some algorithm updates have been made. However, the app is now superpowered! The backend is also hosting an efficient Tensorflow-Node model that does detection on-server, so devices don't have to sacrifice processing power and time on their devices! Still, if you'd like, you can switch to the in-browser option with a simple toggle ;)
- Gif exports are now supported, although since the compilation and encoding of all outputs is done in-browser, .gif encoding is still a little slow. Bear with me until I switch to a possible WebAssembly solution. You can switch between .webm & .gif exports with a toggle, too!
- Created media is auto-hosted on a Google Cloud Storage solution, and seperated with unique urls. Also, you get a QR code & link for your creation, that can be shared publicly, and viewed without logging in!
- Montages can have custom-colored backgrounds, made possible with a Color Picker in the Editor page. Lastly, improved UI/UX design and a fresh new dashboard are sure to make it worth checking out!
That's all the updates for now, folks. Hope to have more, soon. Before I forget, here's a new write-up for the updated project. If you can, any reactions or comments there would be very welcome - https://dev.to/gulliblelamb/riffr-create-photo-montages-in-the-browser-with-some-ml-magic-5271
Cheers, Palash
Log in or sign up for Devpost to join the conversation.