Panorama stitching
Combine overlapping images through matching, camera estimation, warping and blending.
Shared content ties neighbouring images into one panorama.
Match overlapping features and estimate camera relationships.
Alignment establishes a common geometry. The orange line illustrates a seam through the overlap.
A successful stitch requires consistent geometry and exposure. The final scene is an illustrative panorama, not a native stitch of the two drawings.
Compensate exposure and blend overlaps across scales.
Illustrative example. The stages explain the method; they are not a live OpenCV execution.
Try it on an image
Experiment at pixel level
Attempt a panorama from two overlapping photos. The synthetic pair has translation only; real panoramas need sufficient texture and overlap. Failure to estimate a panorama is reported explicitly.
The engine loads on your first run. Your images stay in this browser.
Scroll over either image to zoom at the pointer. Use the scrollbars to pan both views over the same relative area. Zoom is relative to the input; pixel coordinates belong to each image. Warps can change scene correspondence.
Pixel inspector RGBA · native values · matched scale · 9 × 9 output pixels
Select a pixel
Select a pixel
When to use it
Create panoramas when the scene and camera motion fit the selected stitching model.
How it works
- 01Match overlapping features and estimate camera relationships.
- 02Warp images onto a common projection and select seams.
- 03Compensate exposure and blend overlaps across scales.
matches → common projection → seams and blending
What to tune
Choose panorama or scan assumptions deliberately. Check the returned status and inspect overlap consistency before trusting the composite.
Where it breaks down
Parallax, moving objects and weak overlap produce ghosts or alignment failures. A stitcher is not a general scene reconstruction system.
TypeScript API
Open an entry for its exact overloads, parameter descriptions, result ownership and pinned upstream source.
All of these calls execute on the CPU. Native objects need explicit disposal. See matrices and ownership and build compatibility.
Related methods
Non-local means denoising
Average pixels whose surrounding patches look similar, even when they are not immediate neighbours.
Image inpainting
Fill a masked region by propagating nearby image structure and colour into it.
Seamless cloning
Blend a region by solving for pixel values that fit selected image gradients and boundary conditions.
HDR merging and tone mapping
Combine exposure information or fuse well-exposed regions, then map the result into a displayable range.