Image Tools — Compress, convert, resize, remove background — without uploading.

Why Pickrack image tools run in your browser

Image processing online has historically meant uploading your photo to a third-party server, waiting, then downloading the result. That model dates from a time when JavaScript engines were too slow to handle decoding and re-encoding images at acceptable speed. Both of those constraints are gone in 2026. Modern browsers expose the Canvas API, Web Workers, and WebAssembly, which together let JavaScript do everything from JPEG re-encoding to neural-network background removal at near-native speed.

Every one of Pickrack's eight image tools runs entirely in your browser. Open DevTools, switch to the Network tab, click any tool button — there are no upload requests. Your photo is loaded into browser memory, processed on your device, and downloaded back. There is no server-side path for these tools. We could not access your image even if we wanted to.

Which Pickrack image tool to use when

• Shrinking a photo before email or upload — use Image Compressor. Adjustable JPG quality, typical 1-3MB photo compresses to 200-400KB without visible loss.
• Changing dimensions for social or web — use Image Resizer. Aspect ratio locked by default, common presets for Instagram, Twitter, LinkedIn, profile photos.
• Converting between formats — use Image Converter. Supports JPG↔PNG↔WebP↔AVIF in any direction. WebP is typically 30% smaller than JPG at equivalent quality; AVIF is even smaller but browser support varies.
• iPhone HEIC photos that won't open on Windows — use HEIC to JPG. Batch up to 50 photos at once. JPG output works everywhere.
• Removing the background of a photo — use Background Remover. Uses @imgly/background-removal, an ONNX runtime WebAssembly model. The first conversion downloads ~30MB of model weights into your browser cache; subsequent conversions are instant.
• Cropping to a specific aspect ratio — use Image Cropper. Preset ratios (1:1 Instagram, 16:9 YouTube, 9:16 stories) or freeform drag handles. PNG or JPG export.
• Enlarging a small image 2-4× — use Image Upscaler. Stepped bicubic resampling. Honest about limits: not an AI super-resolution model (Real-ESRGAN-style), so won't hallucinate missing detail.
• Extracting brand colors from an image — use Color Palette Extractor. Median-cut quantization. Adjustable 3-12 swatches with hex/RGB/CSS-array output.

EXIF, privacy, and what compression actually does

Canvas-based image processing has a side-effect that's worth being explicit about: EXIF metadata is stripped. Camera model, GPS coordinates, capture timestamp, lens information — all gone in the output. This is almost always a privacy benefit. Photos shared online stop leaking your home address. Photos posted to dating apps stop revealing the time you took them. If you want to preserve EXIF (for example, when archiving photos by date), use a desktop tool like exiftool to copy metadata back from the original onto the compressed output.

Image compression here uses the browser's standard JPEG/WebP/PNG encoders. They are not as aggressive as paid tools like TinyPNG (which runs custom MozJPEG with smarter quantization) — typical Pickrack compression is 10-20% larger than TinyPNG output for the same visual quality. The tradeoff: TinyPNG uploads your photo, Pickrack does not. For most people most of the time, that tradeoff is worth it.