Image Resizer
Resize to exact pixels, social media presets, or print dimensions — in your browser.
100% Private
Drop or tap to resize images
Fast. Private. No uploads.
JPEG · PNG · WebP · HEIC — up to 50MB each
Output preview
Upload an image
to preview
to preview
1280 × 720 px
Ratio: 16:9
Runs entirely in your browser. No uploads. Your files stay private.
Resize Images In The Browser With Canvas And Aspect-Ratio Math
Image Resizer changes pixel dimensions using the Canvas 2D API. Each image is decoded into an HTMLImageElement, drawn into a fresh Canvas at the target width and height, and re-encoded with toBlob in your chosen format. The Canvas drawImage call uses the browser's native bilinear interpolation for downscaling and bicubic-style smoothing for moderate upscaling, which is the same path used by macOS Preview and Windows Photos.
Five resize modes cover what people actually need. Exact accepts width and height directly, with an aspect-ratio lock that recalculates the other axis as you type. Width-only and Height-only resize one axis and let the other scale proportionally. Percentage scales both axes uniformly. Cover crops to fill the exact target dimensions (the same algorithm CSS object-fit: cover uses) with a draggable preview so you choose what to keep.
Preset libraries cover the dimensions different platforms actually want. Social media presets ship with 14 entries for Instagram feed and stories, Facebook posts, X header and post images, LinkedIn banner, TikTok vertical, YouTube thumbnail, and Pinterest pin sizes. Print presets target A4, US Letter, A5, postcard, and business card sizes at 300 dpi, the resolution every commercial printer in the world expects.
Format conversion happens during resize at no extra cost. Pick JPEG, PNG, WebP, or AVIF as the output and the Canvas re-encode handles both the resize and the format change in a single pass — no need to chain Image Resizer into Image Converter. Quality controls apply to the lossy formats; PNG is lossless and ignores the slider.
Up-scaling with Canvas interpolation cannot invent detail; it merely smooths what is already there. For genuine super-resolution upscaling, you need a model like Real-ESRGAN. For everyday work — making a 4K phone photo fit a 1200-pixel blog hero — bilinear downscaling is exactly the right tool and produces sharp results.
Batch resize is supported by queueing files through the same pipeline sequentially. Up to a few dozen images can be processed in one go on desktop browsers; mobile browsers cap at fewer because of memory pressure. Outputs can be downloaded individually or as a single ZIP.
Everything happens locally. Source files live as object URLs in your tab, the Canvas exists only in the page, and the encoded Blob is downloaded directly. There is no server-side resizing service behind the scenes, which is why the tool can promise that your images never leave your device.
Common Use Cases
01
Website and blog hero images
Resize 4K phone photos down to 1600 px wide so the CMS uploads stay light and the Largest Contentful Paint metric improves.
02
Social media platform compliance
Hit the exact required dimensions for Instagram feed (1080x1080), TikTok (1080x1920), or LinkedIn banners (1584x396) without cropping by eye.
03
Email and presentation embedding
Down-scale large photos to roughly 1280 px so they embed cleanly in slides and email attachments under common 25 MB caps.
04
Ecommerce marketplace photos
Resize to Amazon's minimum 1000 px, Etsy's 2000 px, or Shopify's 2048 px requirements before uploading product photos.
Frequently Asked Questions
Downscaling preserves quality well — bilinear interpolation in Canvas is what most native preview apps use. Upscaling smooths what is already there but cannot invent detail; for genuine super-resolution, use an AI upscaler. Always start from the highest-resolution source you have.
Yes. Use Width-only or Height-only mode and the other axis scales automatically. In Exact mode there is a lock toggle so typing in one field updates the other. The aspect ratio is stored as a precise float, not a rounded percentage.
Cover crops the image to fill the target dimensions exactly — the same algorithm CSS object-fit: cover uses. The preview is draggable so you can choose which part of the image to keep.
JPEG, PNG, WebP, AVIF, GIF, BMP, and any other format the browser can decode in an HTMLImageElement. HEIC from iPhones is not handled here — convert with HEIC to JPEG first.
JPEG, PNG, WebP, and AVIF. AVIF availability depends on your browser; older Safari versions can decode AVIF but cannot encode it via Canvas, so the tool falls back to WebP in those cases.
Files up to 50 MB are accepted. The actual ceiling depends on device memory; mobile browsers cap each tab tighter than desktops, so very high-resolution photos may need to be downscaled in two passes on phones.
All print presets are sized at 300 dpi, the standard for commercial printing. A4 maps to 2480x3508 pixels at 300 dpi, US Letter to 2550x3300, and so on. The output PNG/JPEG carries the right pixel dimensions; some print pipelines also expect the dpi metadata to match, which Canvas does not write — set it in your printer's PDF export if needed.
No. The decode, the resize, and the encode all happen in a Canvas inside your tab. The exported Blob is downloaded directly. No image bytes leave the browser.
Yes. Drop several files in, set the target dimensions and format, and the tool processes them sequentially. Outputs download individually or as a ZIP.
Canvas re-encoding strips EXIF, IPTC, and XMP metadata. The orientation tag is honoured before drawing, so the output displays the right way up, but the metadata blocks are gone from the new file.
Step-by-step guide
How to resize an image online
Walk through every step with screenshots, format-specific tips, and the platform-by-platform limits you need to know.
Advertisement