What a mosaic workflow is
A mosaic workflow combines several overlapping images into one larger result. Each panel shows only part of the scene, so the software has to determine how neighboring panels fit together.
How overlapping panels are used
Overlap is what makes stitching possible. When neighboring panels share enough recognizable detail, CieloStitch can estimate how they align in one combined layout. If overlap is weak or inconsistent, matching becomes harder and the result can become unstable.
Main stages of stitching
- Input review: Load the panel set and remove or exclude any frames that clearly do not belong.
- Engine selection: Choose whether the Simple engine or the full Cielo engine is the right starting point for the job.
- Matching: Find shared features between overlapping images.
- Alignment: Place the panels into one common mosaic layout using Translation, Affine, Homography, or APAP depending on the capture.
- Blending: Merge transitions so panel boundaries are less distracting, with ghost-aware seam handling available for difficult overlaps.
- Output: Export the final stitched image.
What changed in newer builds
- Simple engine: A lightweight OpenCV-based path is now available for quicker runs and previews.
- Cylindrical projection: A projection path for wide horizontal sweeps can be enabled when planar geometry starts to drift.
- APAP: A locally varying warp is available for harder perspective or parallax cases.
- Ghost-aware seams: Riskier overlaps can route through a content-aware seam path instead of a generic fallback.
Why consistency matter
The consistently captured image data matters. Panels that differ too much in scale, exposure, orientation, or sharpness are harder to reconcile into one clean result.
What affects output quality
- Strong overlap between panels
- Consistent capture quality across the session
- Enough subject detail to match reliably
- Realistic expectations for the source data
Typical workflow mistakes
- Using panels with too little overlap.
- Mixing captures that differ too much in scale or orientation.
- Using cylindrical projection or APAP before checking whether a simpler baseline already works cleanly.
- Expecting weak source frames to become a strong final mosaic automatically.
- Changing many advanced settings before checking whether a simple baseline stitch works.