Resize and convert an image online
This free Resize Image Converter changes pixel dimensions, crops or frames an image, converts its format, and lets you compare the actual original and output file sizes before downloading. Choose one image for precise control or process up to 20 images into a ZIP. The work stays in your browser: the tool does not upload image files to CalculatorGeek.
For a proportional resize, enter one dimension and keep the aspect-ratio lock on. For an exact 1200 x 630 or 1080 x 1080 canvas, choose Crop to fill when trimming is acceptable or Fit to frame when every part of the source must remain visible.
For the next related calculation, use the Convert Image to JPG. Convert browser-decodable images to JPG in a batch with transparency color, maximum dimension and explicit quality controls.
How to resize an image without distortion
- Choose an image. The converter reads its decoded pixel dimensions and file size locally.
- Set the output box. Enter pixels, a percentage, or physical units with a pixels-per-inch value.
- Choose the sizing method. Fit preserves everything, crop fills the exact box, frame adds space, and stretch changes proportions.
- Choose a format and quality. PNG is lossless; JPG, WebP and supported AVIF output use the quality control.
- Inspect and download. Confirm the rendered dimensions, encoded format and actual byte size before saving the file.
After resizing, use Compress Image to tune the final encoded file size, or choose Crop Image when the source rectangle needs precise coordinates.
To compare this result with another relevant method, open the Rotate Image. Rotate by any angle from -180 to 180 degrees, flip either axis, and package multiple processed images in a ZIP.
Fit, crop, frame or stretch: which mode should you use?
| Mode | Output dimensions | What happens | Best use |
|---|---|---|---|
| Fit within | May be smaller than the box | Keeps the full image and its proportions | General resizing without cropping |
| Crop to fill | Exact box | Scales and removes equal excess from the center | Thumbnails and fixed slots |
| Fit to frame | Exact box | Keeps the full image and adds the selected background color | Products, documents and logos |
| Stretch exact | Exact box | Forces width and height independently | Only when changed proportions are intentional |
Aspect-ratio lock prevents accidental distortion. Enlargement is off by default because creating more pixels cannot recover detail that is absent from the source.
Image formats, quality and file size
JPG is useful for photographs and replaces transparency with the selected background. PNG preserves transparency and uses lossless canvas output. WebP often provides a smaller web-ready file while retaining transparency. AVIF is offered only when the browser can encode it. In Auto mode the converter preserves JPG, PNG, WebP or AVIF where supported; GIF and BMP inputs export as PNG.
The output panel reports measured encoded bytes, not an estimated saving. A higher quality value does not guarantee a larger file for every image, so compare the preview and actual size. Continue with Compress Image when file-size reduction is the primary task.
Worked image resizing examples
Preserve a 4:3 photo
A 4000 x 3000 image fitted inside 1200 x 900 scales by 0.3 in both directions, so the output is exactly 1200 x 900. No crop or padding is needed.
Fill a 1200 x 630 link-preview box
The same 4000 x 3000 image has a 4:3 shape, while 1200 x 630 is wider. Crop to fill scales to 1200 pixels wide and trims equal vertical excess from the center. Fit to frame instead keeps the entire photo and adds space at the sides or top and bottom as required.
Reduce dimensions by half
An 800 x 600 image set to 50 percent becomes 400 x 300. The pixel count falls from 480,000 to 120,000, which is one quarter of the original pixel count, although the encoded file-size change depends on format, image detail and quality.
Inputs, outputs and result meaning
Source width and Source height are the decoded dimensions of the selected file. Target width and Target height define the requested box after the selected unit is converted to pixels. The sizing method decides whether that box is a boundary or an exact canvas.
Requested output pixels is target width multiplied by target height for an exact crop, frame or stretch. Fit within can produce fewer pixels because it preserves the image proportions inside the box. The visible result also reports the actual rendered width, rendered height, encoded format, output byte size and percentage file-size change; those measured values are the ones to use when deciding whether to download.
Bulk image conversion and privacy
Bulk mode applies one format, target box, sizing method and quality value to as many as 20 images, processes them sequentially in the browser, and packages successful outputs into a ZIP. Individual failures remain visible instead of being hidden by a partial download.
Selected files are decoded into browser memory and temporary object URLs. CalculatorGeek does not receive the image bytes, and replacing the image or leaving the page releases the active preview URLs. Browser memory and device performance still set practical limits, so the public caps are 25 MB and 80 megapixels per input and 64 megapixels per output.
Image SEO and accessible metadata
Useful image search optimization starts with a descriptive filename, relevant surrounding text and alternative text that explains the image's purpose. The metadata helper turns your own subject and page context into a draft; it does not inspect pixels or add invented details. Decorative images should normally use an empty alt="" value, while meaningful charts, screenshots and product images need text that communicates the information a sighted reader receives.
Resizing can reduce unnecessary transfer when a page was serving images much larger than their rendered size, but no individual format, filename or alt-text pattern guarantees rankings or AI citations. Match the image to the page's actual topic and verify that the downloaded version still looks clear at its intended display size.
Pixels, physical units and browser limits
Digital layouts are controlled by pixel width and height. Inches, centimeters and millimeters in this converter are convenience inputs: they are multiplied by the selected pixels-per-inch value to calculate pixels. For example, 4 inches at 300 pixels per inch becomes 1200 pixels.
That conversion does not embed a custom print-density tag. MDN documents that canvas image output uses 96 dpi resolution metadata where the format supports it. Print workflows that require a specific embedded DPI, color profile, multi-frame animation or retained EXIF metadata need a dedicated desktop editor after the pixel resize.
Frequently asked questions
Is this Resize Image Converter free?
Yes. The current browser tool has no account requirement, watermark or paid conversion tier.
Are my uploaded images stored?
No image bytes are uploaded to CalculatorGeek. Processing occurs locally in the current browser session.
Can I resize multiple images at once?
Yes. Bulk mode processes up to 20 supported images and downloads successful results in one ZIP.
What is the difference between resizing and compressing?
Resizing changes pixel dimensions. Compression changes how efficiently pixel information is encoded. This tool can do both in one output.
How do I resize without cropping?
Use Fit within to keep the whole image with no padding, or Fit to frame to keep the whole image inside an exact-size canvas.
Does converting a GIF preserve animation?
No. Animated inputs export the decoded still frame. Use an animation-aware editor when every frame must be retained.
Which format is best for a website?
It depends on the content and browser requirements. JPG suits many photographs, PNG preserves lossless transparency, and WebP can be efficient for photos or transparent graphics. Test the actual output rather than relying on one format rule.
Sources and method transparency
The image pipeline uses the browser's Canvas drawImage operation for scaling and cropping and canvas toBlob for the downloaded encoding. Metadata guidance follows Google Images SEO best practices, and the performance explanation is bounded by web.dev responsive-image guidance. Sources and browser behavior were reviewed on October 14, 2026.