Stereo Anaglyph Maker
Combine an existing left-eye and right-eye photograph into a red–cyan anaglyph. Place the right image in the left image’s coordinates using integer X/Y offsets and an extra horizontal disparity. Only pixels covered by both eyes are composed: crop to that overlap, or retain the left canvas with uncovered pixels transparent. The result uses left red and right green/blue channels, so red–cyan glasses with red over the left eye are the intended viewing convention. This tool does not infer depth, automatically align viewpoints or turn one photo into a true stereo pair.
Key features
- Import separate still PNG, JPEG or WebP photographs for the left and right eyes
- Align with integer X/Y offsets and a separate horizontal disparity adjustment
- Inspect the exact matching overlap in side-by-side left/right previews
- Crop to the overlap or keep the left canvas with transparent uncovered pixels; add a black border
- Export the red-left/cyan-right PNG and a JSON record of the alignment
How to use
- Load a left-eye and a right-eye photo, or open the sample pair; swap eyes if they were assigned in reverse.
- Adjust X/Y alignment and horizontal disparity until the shared region matches your intended depth.
- Choose a cropped overlap or a left-sized canvas with transparent nonoverlap, and set an optional black border.
- Generate the anaglyph and compare the two aligned-eye previews with the red–cyan result.
- Download the PNG and settings JSON; view the PNG with red on the left eye and cyan on the right.
Use cases
- Convert a two-camera stereo pair into a red–cyan viewing image
- Tune a horizontal disparity before sharing a simple 3D photo
- Inspect where a handheld stereo pair still overlaps after manual alignment
- Keep transparent margins for later compositing in a graphics editor
Frequently asked questions
Which color goes to each eye?
The output red channel comes from the left photo. Green and blue come from the right photo. Use glasses with red over the left eye and cyan over the right.
What do positive offsets and disparity mean?
A positive X or disparity moves the right photo to the right in left-photo coordinates. Positive Y moves it down. The effective horizontal move is X plus disparity, in whole pixels.
What happens to areas seen by only one camera?
They are never synthesized from a single eye. Overlap crop removes them. Left-canvas mode keeps the original left size and leaves them transparent.
Can this make 3D from one photo or align automatically?
No. Supply two different viewpoints and adjust their translation yourself. This tool does not estimate depth, rotate, rescale or rectify lenses.
How are transparency and color handled?
Each input pixel is composited over white before taking the channels. The calculation uses the browser-decoded 8-bit sRGB values and exports an opaque overlap in PNG; color-profile and EXIF metadata are not copied.
Privacy
The photos and output are processed in browser memory. This tool sends no photos to a server and does not automatically save settings to browser storage. PNG and JSON files are saved to your device only when you download them.
Comments & questions