How offline image cropping & resizing works
This tool uses your browser's Canvas API — no external libraries, no WebAssembly, no server. When you load an image, it's decoded natively into a canvas. The crop selection mask is rendered as HTML overlays with CSS. Final export uses canvas.toBlob() with your chosen format and quality. At no point does the file leave your device.
Crop + resize: two tools in one
Most online tools force you to choose: crop OR resize. This combines both into a single workflow:
- Crop tab — select a region of your image with click-and-drag or the 8-point handle system. Lock to any aspect ratio (1:1 for Instagram, 16:9 for YouTube thumbnails, etc.).
- Resize tab — set exact pixel dimensions. Enable aspect ratio lock to auto-calculate the matching dimension. Works on the cropped region — crop first, resize after.
Social media presets
Every preset is included — no need to look up dimensions:
| Platform | Preset | Dimensions |
|---|---|---|
| Post | 1080 × 1080 | |
| Story / Reel | 1080 × 1920 | |
| Link Post | 1200 × 630 | |
| Twitter / X | Post | 1200 × 675 |
| YouTube | Thumbnail | 1280 × 720 |
| Post | 1200 × 627 | |
| Pin | 1000 × 1500 | |
| TikTok | Post | 1080 × 1920 |
Output formats: WebP, AVIF, JPEG, PNG
Choose your output format in the sidebar:
- Keep original — matches your input format (JPEG → JPEG, PNG → PNG, etc.)
- WebP — Google's modern format. 25-35% smaller than JPEG at the same quality. Supported by all modern browsers.
- AVIF — next-generation format. Even better compression than WebP. Available in Chrome 123+ and Firefox 133+. The tool auto-detects browser support.
- JPEG — universal compatibility. Adjustable quality slider.
- PNG — lossless. Largest files but zero quality loss. No quality slider needed.
Privacy: your files, your machine
Most image resizers upload your photos to a cloud server for processing. That means:
- Your original, uncropped image sits on someone's disk
- Any faces, locations, or documents in the image are visible to the server
- EXIF metadata (GPS, camera, timestamp) is readable by the operator
- You're trusting their deletion policy, security practices, and third-party integrations
PrivateConvert eliminates this entire attack surface. The crop overlay is HTML. The resize math is JavaScript. The rendering is Canvas API. All of it runs inside your browser's sandbox — at no point does any image data travel over the network.
About EXIF/metadata stripping
All canvas operations inherently strip image metadata. When your browser draws an image to a canvas element and exports it via toBlob(), the resulting file contains only pixel data — no EXIF, no GPS coordinates, no camera model, no timestamp, no color profile metadata. This is not something we toggle on/off; it's a fundamental property of how the Canvas API works. For privacy-focused users, this is a feature — your output image is anonymous by default.