
Closed
Posted
Paid on delivery
I’m building a small desktop tool for Windows that lets users drop in any portrait and instantly receive a perfectly aligned passport-size JPEG. The key is accuracy and speed, so I need robust AI face detection that automatically centres the head, levels the eyes, and corrects shoulder tilt without manual tweaking. Once the crop and alignment are finished, the file should be saved straight to JPEG at the standard passport dimensions I’ll supply. While the core focus is the one-click crop, I’d like the codebase structured so secondary edits—brightness or contrast tweaks, red-eye fixes, even full background removal—could be added later without rewriting the detection workflow. If you already have modular models or libraries that handle those, let me know; otherwise, neat hooks or an extension interface will be enough. Deliverables • Windows installer (or portable EXE) ready to run on current versions of Windows • Source code with clear comments and a short README that explains model training or external AI services used • A quick user guide illustrating the drag-and-drop flow, the AI alignment in action, and the final JPEG export path I’ll test by running a batch of sample portraits with varying orientations; every output must respect the official passport frame and appear level to the naked eye. Let me know which frameworks or AI libraries you plan to leverage, and your timeline to a working prototype.
Project ID: 40511996
13 proposals
Remote project
Active 57 yrs ago
Set your budget and timeframe
Get paid for your work
Outline your proposal
It's free to sign up and bid on jobs