Photo Merge (HDR, Panorama, HDR Panorama)

crates/merge (layer L3, no UI dependencies) implements the algorithms; crates/engine/src/merge.rs plans, runs and finishes merges in the library; the desktop UI shows a dialog with a live preview. All algorithms are our own implementations from published papers (cited in the module docs).

Using it

• App: select the photos, then Photo Merge ▸ HDR… (Ctrl+H), Panorama… (Ctrl+M) or HDR Panorama… from the photo context menu (or the dialog.merge* commands). The dialog previews the merge at ≤ 1024 px and re-runs the preview when an option changes; as in a professional photo organizer, a click on the preview zooms in on that point (drag to pan, click again to fit), and it stays there while options change; Merge runs the full merge in the background (a row in the activity stack shows its progress, with ✕ to stop it). The result is written next to the first photo as <name>-HDR.dng, -Pano.dng or -HDR-Pano.dng, imported and selected. Photo Merge ▸ HDR for Each Bracket Set merges many brackets in one go with the last HDR options: one HDR per stack among the selected photos, then per bracket (photos taken less than a second apart, after the shutter time, at exposures at least ½ EV apart; a burst at one exposure is skipped), one after the other in the activity stack.
• Commands (CLI, MCP, control channel):
  • merge.hdr {ids?, align=true, deghost=none|low|medium|high, autoSettings=true, stack=false, preview=false, showOverlay=false, previewPath?}
  • merge.panorama {ids?, projection=auto|spherical|cylindrical|perspective, boundaryWarp=0..100, autoCrop=false, fillEdges=false, autoSettings=true, maxMegapixels (default: the photos' own resolution), preview, previewPath?}
  • merge.hdrPanorama {ids?, bracket=0 (from EXIF), …both sets of options}
  • merge.hdrBatch {ids?, align, deghost, autoSettings, stack} → {merged: [{id, path, …}], failed: [{photos, error}]}
• CLI: imago-cli merge hdr|panorama|hdr-panorama|hdr-batch [options] FILES… (hdr-batch finds the brackets among the files, e.g. a whole folder); imago-cli synth-merge hdr|panorama -o DIR writes synthetic inputs (procedural scene).

What it does

HDR (hdr.rs): frames ordered by EXIF exposure (or median luminance); the middle one is the reference. Auto Align: features on exposure-normalised luminance, RANSAC translation / similarity / homography (the simplest model that explains ≥ 97 % of the matches). Frames are resampled with Lanczos-3, held to the range of their 4 × 4 neighbours so edges don't ring (imago_raster::sample); bilinear lost a third of the finest detail. Exposure ratios are measured (median log ratio of well-exposed pixels between brightness neighbours), EXIF is the fallback. Radiance is a weighted mean in linear light, weight k · (1 − smoothstep(0.8·clip, 0.96·clip, max channel within 1 px)): a pixel next to clipped photosites was demosaiced from them and reads low in the clipped channel, so it goes with them (on real ARW brackets the brightest frame's green was 3–5 % low there, and bright windows took a magenta tinge). The frames are taken in brightness order (ties by content), so the result doesn't depend on the order they come in. Where the reference is at or near its clip the scene is at least that bright, so a frame that reads far darker there (an object moved: a star, the moon, a passing light) is left out and the clipped highlight stays (always on: a moving star otherwise left a black hole). Deghost: a per-pixel consistency test on ≤ 2048 px versions against the reference radiance (with a noise floor; the reference predicts down to 0.1 % of its clip, so a brighter frame's ghost in a night sky is caught), dilated and feathered masks; the union is the deghost overlay. Output: 16-bit float DNG. Lens corrections: frames are merged uncorrected, and the DNG's OpcodeList3 carries the reference frame's (WarpRectilinear, WarpRectilinear2, FixVignetteRadial: a DNG's own, or a camera's correction a raw reader converts, e.g. Sony, Panasonic), re-expressed for its default crop (Opcode::cropped). So Optics ▸ Enable Lens Corrections works on the result as on the sources; a vertical bracket keeps its pixels unturned and its corrections.

Panorama (pano/): MOPS-style features, pairwise RANSAC homographies with Brown & Lowe's verification, focal from homographies (or EXIF 35 mm), rotations chained along a maximum spanning tree, Levenberg–Marquardt bundle adjustment of rotations + focal lengths, straightening. Projections: spherical, cylindrical, perspective; Auto picks perspective for compact fields of view < 110°, cylindrical when the vertical field is < 90°, else spherical. Exposure: EXIF, then gain compensation from the overlaps. Seams: each pixel goes to the image with the highest centre-weighted feather weight; multi-band blending (Laplacian pyramids) in log space. Boundary Warp: a separable stretch of the smoothed valid-area edges to the canvas edges. Auto Crop: the largest rectangle of image data (set as the develop crop, so it stays adjustable). Fill Edges: push–pull diffusion. Output: 16-bit integer DNG (16-bit float for HDR panoramas).

DNG output (output.rs): LinearRaw, 3 samples, Deflate tiles; values scaled below 1.0 with the scale in BaselineExposure; raw sources keep their DNG colour tags (camera RGB), other sources get ColorMatrix1/ForwardMatrix1 of their RGB space.

Measured (synthetic tests, cargo test -p imago-merge)

• HDR, 3 brackets ±2 EV with 14-bit noise and handheld shifts: alignment error < 0.3 px, exposure error < 0.02 EV, radiance median log error < 0.03 (noise-limited, bias < 0.01), recovered highlights < 0.05; moving object: ghost error 0.2–2.0 without deghosting, < 0.06 with High; a star that moves across a night sky keeps its clipped core and, with High, leaves no ghost of the brightest frame.
• Real brackets (Sony A7R VI, 67 MP, 2026-10-14): frames agree with the merged radiance within ~1 % from 3 to 90 % of their clip in all three channels; measured exposure steps match EXIF within 0.03 EV.
• Panorama, 5 views over 182° with a 1.3× exposure drift: relative rotations < 0.1°, focal < 1 %, BA RMS ≈ 0.35 px; median log error vs ground truth < 0.02; no seam steps.

Limits (known gaps)

• No lens-distortion model in the panorama bundle adjustment (rectilinear lenses assumed); no 360° wrap-around; flat (translating-camera) scans are fitted with the rotation model. Panoramas and HDR panoramas carry no lens corrections, nor does an HDR whose frames don't share an orientation (they are turned before merging).
• No camera response calibration for JPEG brackets (sRGB decoding is taken as linear light).
• AI Denoise doesn't run on merged DNGs (they are linear RGB; the denoiser takes Bayer raws), unlike a professional photo organizer's.
• Boundary Warp is a separable stretch, not a content-preserving mesh warp; Fill Edges is diffusion, not patch synthesis.
• Create Stack: the result goes on top of a collapsed stack of its sources (Catalog::stack_with_ops); its caption also lists them.
• Panorama output is limited to maxMegapixels (default 40) — blending memory is ~25 bytes/pixel.
• HDR memory: every frame is held as 32-bit float RGB (~0.86 GB per 67 MP frame at the peak, plus ~1.2 GB); full-size photos are decoded one at a time. Measured on 67 MP raws: 3 frames 3.8 GB, 4 frames 4.7 GB, 5 frames 5.5 GB (was 4.8 / 6.3 GB for 3 / 4). Half-float frames would halve the per-frame share.
• The browser build has no file access for merges yet.