Project Case Study

PhotoSahi

Privacy-first passport and visa photo generation running entirely inside the browser.

By Mantosh Kumar · Updated July 13, 2026

Try PhotoSahi GitHub

Problem and approach

Official photo requirements vary, while many tools ask people to upload sensitive images. PhotoSahi uses HTML5 Canvas to crop, resize, adjust, and export photos locally in the browser—without a backend or mandatory upload.

Architecture

One private, browser-only journey from the original photo to a ready-to-use file.

Your photo stays on this device No backend, mandatory upload, or server-side image storage.
  1. 1 · Input Choose a photo Read locally from your device
  2. 2 · Browser Process with Canvas Crop, resize, and adjust
  3. 3 · Prepare Build the output Apply the selected photo preset
  4. 4 · Output Download the result Save the processed file directly

Device → browser memory → download. The image never needs to cross a network boundary.

Engineering Decisions

Browser-only processing

Local canvas operations avoid mandatory uploads, backend latency, server cost, and an unnecessary trust boundary.

Performance

Canvas operations keep common transformations immediate across desktop and mobile browsers.

Privacy and trade-offs

Local processing improves privacy, but device capability and browser compatibility become product constraints.

Next improvements: verify more document presets, refine background controls and previews, and improve first-use guidance.