Skip to content

Combine images side by side or vertically

Stitch before-and-after shots, mobile app screenshots, or comic panels side by side or vertically with clean alignment and zero server uploads.

Your files are open only in this tab. Nothing is uploaded.

Images to combine (2–10)

Stitches screenshots, before/after comparisons, or photo strips side by side or vertically.

Drop 2 to 10 images here

Reorder panels and adjust scaling locally without uploading.

How to use Combine images

  1. Add 2 to 10 images or load the sample pair. Drop PNG, JPG, or WebP shots into Add images, or test layout rules with the built-in before-and-after sample images.
  2. Reorder panels and choose a stitch direction. Use the Up, Down, and Remove buttons on up to 10 panels, then select Side by side (Horizontal) or Stacked (Vertical).
  3. Tune dimension matching, alignment, gaps, and canvas fill. Pick Match smallest or Keep original sizes, set Center, Start, or End alignment, and adjust Gap between images or Outer padding.
  4. Export your stitched PNG, JPG, or WebP strip. Check the live combined preview dimensions and download the finished strip offline.

Your files are processed on this device and never uploaded. They stay private, and there's nothing to delete afterwards.

How does dimension matching prevent aspect distortion?

Proportional dimension matching scales uneven photos along their shared axis so the other axis adjusts at the exact source ratio instead of stretching faces or text. Decoding via the MDN Window.createImageBitmap() specification reads real image bounds locally before merging shots offline with zero server uploads.

When teams combine images from before-and-after screenshots or mixed portrait-and-landscape mobile frames, naive canvas tools squeeze rectangles to identical bounds and ruin proportions. By locking the aspect ratio during resizing, every panel keeps natural geometry cleanly inside 8 px gaps and 12 px outer margins.

Every uploaded picture keeps native orientation metadata decoded upfront so tall mobile shots and wide desktop captures share balanced margins. Previewing combined dimensions live makes layout spacing immediately clear before saving local files.

Important: Keeping Keep original sizes enabled avoids resampling altogether. Shorter frames stay at full 1:1 resolution and align vertically or horizontally using Center, Start, or End rules.

Sideways smartphone photos sometimes retain EXIF rotation flags that show rotated dimensions once pixel buffers decode. Fix misaligned captures upfront when you rotate sideways screenshots upright before stitching so every horizontal strip reads level.

Configuring layout direction, spacing, and transparency

Two primary orientations organize panel flow cleanly while worker spacing controls frame every border. Before dropping mixed browser downloads onto the strip, run browser-saved JFIF to JPG normalization if older screenshots carry unrecognized extensions.

Horizontal strips versus vertical scroll stacks

Selecting Side by side (Horizontal) places panels left to right along a single row, matching cross-axis heights when dimension matching is active. Switching to Stacked (Vertical) orders panels top to bottom in a scrolling column, matching panel widths instead.

Clicking Load sample pair inserts a 320 px by 240 px before frame next to a 280 px by 220 px after frame so you can watch layout math update immediately. Panel rows include dedicated Up, Down, and Remove actions for instant sequencing adjustments.

Gap, outer padding, and Porter-Duff alpha blending

Two numeric sliders adjust Gap between images and Outer padding from 0 px to 64 px in crisp 2 px steps, defaulting to an 8 px inner gutter and 12 px frame margin. Unchecking Transparent canvas fills margins with #ffffff or any custom hex backdrop.

Overlay pixels blend onto the background using standard source-over alpha math defined by the W3C Compositing and Blending Level 1 source-over specification. Straight alpha blending computes blended channels accurately without darkening semi-transparent screenshot drop shadows against white or transparent canvases.

Safe 8192 px canvas clamping for high-DPI photos

Combining ten 4K captures side by side can demand tens of thousands of pixels and exhaust browser tab allocations. Whenever total width or height exceeds MAX_OUTPUT_SIDE at 8192 px, the worker applies a uniform proportional downscale factor across all panels, inner gaps, and outer margins.

  • Predictable output bounds: Even extreme multi-panel stacks stay bounded below 8192 px on the longest side to prevent silent GPU canvas failure.
  • Preserved spacing ratio: Gaps and outer padding shrink by the identical ratio as image frames, keeping border proportions intact.
  • Exact copy bypass: Smaller pairs below 8192 px that share matching heights never invoke bilinear resizing or clamping math.

Dimension matching, alignment, and format tradeoffs

Choosing the right scaling mode prevents blurry UI captures while controlling overall export dimensions. Re-encoding to lossy JPEG or WebP uses the browser MDN HTMLCanvasElement.toBlob() method at a fixed 92% quality target.

Dimension matching mode Cross-axis rule Upscaling risk Resolution effect Ideal stitch workflow
Match smallest Shrinks taller or wider frames to shortest panel None (downscales only) Compact, ultra-sharp text edges Mixing high-DPI phone and desktop screenshots
Match largest Expands smaller frames up to largest panel Can soften low-res screenshots Uniform full-size strip Enlarging smaller thumbnails to match a hero frame
Match first image Scales frames 2 through 10 to match panel 1 Depends on panel 1 size Predictable anchor scale Setting a strict template width via the first upload
Keep original sizes Zero scaling; aligns unscaled panels via Alignment None (1:1 native pixels) Mixed edges framed by backdrop Pixel-accurate UI sprites and side-by-side audits

Product teams assembling comparative badge charts sometimes unpack software asset icons alongside screenshots beforehand. Dedicated multi-size ICO icon extraction to PNG ensures every favicon or app icon decodes cleanly before lining frames up side by side.

Stitching crisp screenshots and before-and-after pairs

Choosing between lossless PNG and compressed JPEG or WebP depends heavily on whether your panels contain razor-sharp text or continuous photography. Notice that exported filenames include output pixel dimensions directly in the format combined-<width>x<height>.<ext> for quick verification.

Choosing PNG, JPG, or WebP for crisp UI text

Screenshots showing dashboard tables, code diffs, or mobile typography look sharpest in PNG (Lossless) following the W3C PNG Third Edition lossless specification. Lossless deflate encoding avoids chroma ringing around thin gray fonts and preserves alpha channels whenever Transparent canvas stays active.

Pick WebP (.webp) when combining five or more high-resolution photographs for blog articles where 92% lossy compression cuts file sizes sharply while preserving alpha transparency. Choose JPEG (.jpg) only with solid Background colors for portrait comparisons shared on legacy messaging platforms.

Panel ordering, privacy hygiene, and 10-image limits

Sequential ordering happens strictly inside browser state so rearranging frames 1 through 10 never re-uploads files or wastes bandwidth. Sensitive receipts, front-and-back ID cards, and unreleased app mockups stay confined inside local browser RAM from upload through Download PNG.

  • Strip limits: Keep individual strips between 2 and 10 panels; split longer comic chapters into numbered batches before archiving.
  • Matrix collages: Linear strips work along one horizontal or vertical axis, so switch to multi-row photo grids ready for print when arranging multi-column albums.

Questions

Should you pick Match smallest or Match largest when combining photos of different sizes?

Choose Match smallest when crispness matters because scaling large captures down avoids inventing soft upscaled pixels. Choose Match largest only when keeping every fine pixel of your highest-resolution screenshot intact is more important than keeping smaller panels razor sharp.

Can you combine transparent PNG screenshots without white margins?

Yes, turning on Transparent canvas clears the canvas background completely and exports full alpha transparency in PNG (Lossless) or WebP formats. When saving JPEG files, leave Transparent canvas unchecked and pick a solid Background color because the JPEG format does not support alpha transparency.

How many images can you stitch together at once, and what is the maximum pixel size?

The uploader lets you stitch between 2 and 10 image panels per run. To stay safely within browser memory budgets, any combined output exceeding 8192 px along its longest edge is automatically scaled down proportionally across panels, gaps, and outer padding.

Why does bilinear resampling run only when source dimensions differ?

When every panel already shares the target height or width, or when Keep original sizes is active, bytes copy directly onto the stitched canvas without interpolation. Bilinear RGBA scaling runs strictly on resized panels so matching frames stay completely untouched.

Do uploaded screenshots or personal ID photos leave my device when combining?

No file bytes, dimensions, or filenames leave your browser at any point. Decoding via createImageBitmap, bilinear panel resizing, alpha compositing, and PNG compression run locally inside your browser worker thread.

Updated