The Automatic palette counts flat art's colours by area (a colour needs
about 1% of the picture, 0.5% at Colors 100) and fits that count with
median cut and area-weighted k-means. A stem, a highlight, or a near
white beside a pale background (within 24 levels, so in one box) was
averaged into a large neighbour; mixed-colour merging and Noise could
then remove what was left.
On the flat route only, after the count is fitted, rare_colours finds
the colours too rare or too close for it and keeps those that are
distinct (at least 2.5 in Oklab x 100 from every palette colour and from
the main colour each stands for), flat (mostly one histogram bin) and
drawn (a 4-connected region of their own unblended pixels of at least 8
pixels, or three of at least 4); a colour lying between two others, as
anti-aliasing mixes them, must also have an inside. At most six, largest
region first. They stay fixed while the other colours are refitted
around them; merge_mixed leaves them alone; denoise_keeping keeps their
regions of 4 pixels or more whatever Noise is (smaller specks still go).
Full Tone, Limited and Automatic's continuous-tone route are unchanged,
and denoise without kept labels is as before.
Methods cited in the code: median cut (Heckbert 1982), k-means (Lloyd
1982), Oklab (Ottosson 2020), salient-colour preservation (Fukuda et al.
2019, doi:10.1117/12.2521607). The rule and its numbers are VectorCraft's
own; Adobe does not document how its palettes are chosen.
Tests (code-drawn): a rare stem or patch over 2-8 flat colours, hard
and anti-aliased at 1x and 2x; near-white highlights on #FFF4E0 kept and
still paths at Noise 16; a 3 px #909090 stem on #808080; scattered
single pixels of the same areas add no colour; denoise_keeping at Noise
0-32.
Measured on 66 synthetic flat logos (22 designs x hard 1x, anti-aliased
1x and 2x; local inputs): every key colour and no spurious one in 66
(was 39); missing key colours 0 (was 43), spurious 0 (was 2), merged
pairs 0 (was 46); rendered dE00 0.348 -> 0.138. Four public-domain scan
crops take the continuous-tone route and are unchanged.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_018tBcdwNhptqtmX32JCGFzB
The public documentation says Image Trace warns about images larger
than 1500 pixels on a side but traces them, only more slowly. VectorCraft
traced them silently, and refused any trace of more than 2,000,000
anchors, which a 4000 x 3000 photo at High Fidelity Photo nearly reaches
(1.9 million on a code-drawn photo) and a noisier one passes.
- imageTrace.make and makeAndExpand return `warnings`: one for an image
over 1500 px on a side (traced, but more slowly; a lower-fidelity
preset or a smaller copy is quicker), one for an image over 64 MP
that was traced from a scaled-down copy. The app shows the first in
the status bar.
- MAX_TRACE_ANCHORS is 8,000,000 natively (2,000,000 on wasm, where a
page has at most 4 GB). Measured through imageTrace.make on
code-drawn photos: about 90 bytes an anchor kept in the document;
7.9 million anchors kept 685 MB with an 876 MB peak (4000 x 3000,
Full Tone at Colors 100, Paths 100, Noise 1, 37 s).
- The refusal keeps its translated advice: raise Noise, lower Paths or
Colors, or pick a lower-fidelity preset.
Tests: a_large_image_traces_with_a_warning (1600 x 40 and 1500 x 8
code-drawn images, make and make-and-expand),
trace_warnings_name_a_large_or_scaled_down_image, a compile-time check
of the limit's range, and a UI test that the status bar shows the note.
Source:
https://helpx.adobe.com/illustrator/using/image-trace-results-optimization.html
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_018tBcdwNhptqtmX32JCGFzB
Raster::new checked its RGBA length with assert_eq!, so bytes of the
wrong length (from a file, a plug-in or a control message) would abort
instead of reporting an error. It now returns Result<Raster, TraceError>
and gives TraceError::Decode naming the byte count and the size (the
existing variant, whose text is already translated, instead of a new
message every language would need). The size product is checked for
overflow.
Raster::pixel indexed the bytes directly; it now gives transparent
black outside the raster, or past the end of a raster whose public
fields were set to a size its bytes don't have (which mosaic reads).
The public signature changes: every caller in the workspace was a test
(trace unit tests, Flat Logo tests and the quality harness), updated to
unwrap. No shipped code built a Raster with new; decode, decode_to_fit
and from_fn always make the right length, and trace_within already
refuses a raster whose fields disagree.
Test (code-drawn): a_raster_of_the_wrong_length_is_an_error_not_a_crash
(short, long, empty, zero-width and overflowing sizes; pixel, trace and
mosaic on a raster whose fields disagree).
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_018tBcdwNhptqtmX32JCGFzB
Place junction clusters at their deepest pixel and prune terminal branches
by significance beyond their maximal discs. Merge overlapping junctions
and split crossings with continuing outer arms, then join degree-two runs.
Bound nodes, branches and simplification work, using checked data access.
Keep whole-blob selection, ink refinement, cap detection and fitting rules.
Test junction degrees, crossings, 24-angle lines, pruning, graph caps and
random-blob determinism. Record improved junction degrees and remaining
spur and extra-junction limitations in the centre-lines measurements.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Co-Authored-By: OpenAI <[email protected]>
Compute exact squared Euclidean distances per blob with separable parabola
lower envelopes. Thin with bounded Guo–Hall active sets and retain diagonal
ends during staircase cleanup. Keep the whole-blob width rule and existing
graph, ink, cap and fitting rules, with checked access in the moved module.
Verify EDT against brute force, active sets against plain rounds, and both
diagonal slopes for length and connectivity. Record the centre-lines rerun
and its remaining limitations without changing feature scores.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Co-Authored-By: OpenAI <[email protected]>
Measure Create Strokes on exact code-drawn scenes in aa and hard variants,
including centre/width error, topology, ends, spurs and painted coverage.
Keep the existing allocator report in one ignored test and document the
current baseline without changing tracing behaviour or dependencies.
Co-Authored-By: Codex <[email protected]>
Co-Authored-By: Claude Opus 5.5 <[email protected]>
A small closed loop whose fit collapsed to a single cubic (fit_closed),
or a piece returning to its own junction with fewer than two fitted
cubics (fit_piece), kept every positioned point as a straight segment.
A looser Paths collapses more loops, so it could give more segments
than a tighter one.
Such a loop now tries the fewest segments that still hold it, each only
if the result keeps an area (more than 0.25 px^2) and stays within
1.5 * max(tolerance, 0.5 px) of the points both ways: the fit at up to
three somewhat tighter tolerances (what a slightly higher Paths gives);
for a closed loop then two cubics split at its start and its farthest
point, then the outline simplified at the tolerance and finer. The
positioned points stay the last resort, as before.
Test (code-drawn): a_higher_paths_never_gives_fewer_segments sweeps
Paths 0/25/50/75/100 over 32 seeded 48 px images of soft blobs plus
pixel noise in Color and Black and White at Noise 1 and 3. Before, six
runs had fewer segments at a higher Paths (e.g. 125 -> 79 from Paths 0
to 25); after, none. One image (seed 16, Black and White: 140 -> 136
from Paths 0 to 25, with or without this change) has another cause and
is left out of the sweep.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_018tBcdwNhptqtmX32JCGFzB
Where the pixel at a sharp corner falls to the other colour, the outline
cuts the corner in two one-pixel steps, and between two straight edges
of three cracks or more subpixel::outline makes those steps one point
(a notch). Whether the edges were long enough was measured on the whole
loop. An edge can run on past a junction in one region's loop while
the region across it, whose loop turns at that junction, sees a shorter
edge: one side made the notch and the other did not, so the shared
piece got different points on each side and the abutting curves left a
gap.
The notch now also requires the three cracks of each edge next to it to
lie in the piece (no junction among them), as the corner-pixel shift
correction above it already does.
Test (code-drawn): noisy_colour_images_leave_no_sample_uncovered traces
20 seeded 36 px images of soft overlapping blobs in four colours plus
pixel noise (Color, 5 colours, Noise 1 and 2, Paths 0/50/100) and needs
every one of five sample points per pixel to lie in some path: 93
uncovered before (seeds 9, 11, 13), 0 after.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_018tBcdwNhptqtmX32JCGFzB
Layers::shift read a crack's sub-pixel shift from its inside pixel,
looking for a thin run on the inside before one on the outside. Where
both sides are thin (a one- or two-pixel sliver beside another, as in
soft gradients and noisy scans) the two regions of the crack got
different shifts, so their shared piece got different points and the
abutting curves left slivers that no path covers.
The shift is now the mean of the readings from both sides,
(s(c) - s(c flipped)) / 2, which is exactly antisymmetric in floating
point (Crack::flipped gives the crack seen from its other side).
Tests (code-drawn images):
- a_crack_between_thin_slivers_has_the_same_shift_from_either_side:
every crack's shift is the exact negative of its flipped crack's, on
a noisy ripple of 1-2 px columns at three thresholds.
- shared_boundaries::thin_slivers_on_both_sides_of_an_edge_leave_nothing_uncovered:
a 48 px noisy ripple in Black and White at thresholds 140/157/170 and
Paths 0/50/100, sampled on the shared-boundary check's 4x4 grid per
pixel: 482-1,228 uncovered points per trace before, 0 after.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_018tBcdwNhptqtmX32JCGFzB
With Method Abutting and Shapes on, a region recognised as a shape had its
outline replaced by the exact shape, but the region across that edge kept
its fitted curve unless it was a hole filled by one colour, and even then
the hole was recognised on its own, with its region's combined hardness.
A shape touching a junction (a third colour, or its own colour at a
corner) or the image's border was replaced on one side only, so abutting
colours parted or overlapped along the shared edge, and a square on the
image's corner could reach past the frame.
Cause: Layers::layer recognised an island with no holes
(`[only] => shapes::recognise`) regardless of what lay around it, and a
hole only by `one_colour_within`, each from its own outline and hardness.
Fix: under Abutting a shape is applied only where both faces take it
(`Layers::shared_shape`): the loop has no junction on it and does not
touch the border, the island has no holes and the hole holds one colour,
and the shape stays inside the image. The shape is recognised from the
outline turned one way and started at its leftmost anchor, with that
loop's own hardness, so the island and the hole it fills get the very same
shape. Overlapping is unchanged.
Test: shapes_keep_abutting_neighbours_watertight traces three code-drawn
anti-aliased colour images (a disc in a square plus a disc across its
edge, shapes on the border, tilted tiles) and checks every outline lies
within 0.001 px of a neighbour and a 4 x 4 per-pixel grid is covered
exactly once. Before: gaps of 0.50, 15.1 and 0.73 px, 183 and 276 points
uncovered or doubled; after: 0 px, 0 points. Isolated shapes are still
recognised.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_018tBcdwNhptqtmX32JCGFzB
Flat Logo found its colours only from solid pixels (a uniform 3 x 3
neighbourhood). Line art on paper has no solid area but the paper's: a
stroke one or two pixels wide never fills one, so its colour was a handful
of pixels among tens of thousands of white and was cut as noise (under
0.1% of the solid pixels). The palette came back empty and trace_logo
returned an empty TraceResult, which imageTrace.make reported as a success
with paths 0, anchors 0, colors 0.
Cause: logo::palette::auto_palette had no way to find a colour that has no
solid area, and logo::pipeline::trace_logo treats an empty palette as
nothing to trace.
Fix: when the solid pixels give no colour but white, auto_palette reads
the colours from the strokes' cores instead (line_art_palette): the opaque
ink pixels at least as far from white as each of their four neighbours,
the middle of each stroke where it is least blended with the paper.
Two guards keep that fallback to line art:
- Ink is a colour at least 48 below white in one channel, and the page
must have ink on at least 1% of its opaque pixels. Light grain and a few
sparse marks on otherwise blank paper stay blank, as plain white does.
- The lighter shades along a stroke's anti-aliased edge are local maxima
of their own; a colour that lies on the line from white to a darker
colour of the palette (within 20) is a blend of that ink with the paper
and is folded into it (fold_fringes), the ink keeping its colour.
The rest of the pipeline (unmixing, the flat-colour check and its NotFlat
refusal) is unchanged, and images with a solid colour other than white
never reach the fallback.
Tests (code-drawn inputs):
- thin_line_art_on_paper_is_traced_not_blank: four 2.4 px dark strokes
and two red marks on white. Before: palette [], no paths. After: dark
and red paths along the middle of every stroke and mark.
- anti_aliased_coloured_lettering_has_only_its_inks: rings, slants and
bars 1.6 px wide in red, blue and green. Without the fold: six colours
besides white; after: exactly the three inks.
- sparse_marks_on_paper_are_not_promoted_to_colours: 24 grey 12 x 2 px
marks of 239, 223 or 191 on a 256 x 256 page (under 1% ink). Without
the guard: palette [white, #efefef]; after: empty palette, no paths.
- grainy_blank_paper_has_no_ink: paper grain 243-255 traces to nothing.
- edge_shades_fold_into_their_ink_and_other_colours_stay (unit).
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_018tBcdwNhptqtmX32JCGFzB
contour::trace_mask built a table of outgoing directions for every
pixel corner in a raster pass and then followed it. It now works out
each turn during the walk from the two pixels ahead of the current
corner: right if the one ahead on the right is outside, straight if the
one ahead on the left is, else left (so at a saddle it still turns
right and diagonal pixels stay apart). Loop starts are found by
scanning the pixels for a top edge on the outline, with one flag per
pixel for a top edge already walked. The walk is bounded by the grid
size, and a mask of the wrong length now gives no loops instead of
failing an assert_eq!.
The output is identical: the_walk_equals_the_stored_direction_walk
compares it with a test-only copy of the previous walk on all 65,536
4x4 masks and 400 seeded 23x17 masks.
Time (best of 7, release, 4 cores): the walk alone on 2000x2000 masks
19.1 vs 26.6 ms (smooth pattern), 81.3 vs 93.2 ms (12% random pixels),
223.1 vs 238.1 ms (50% random); whole 1000x1000 traces 73.1 vs 77.1 ms
(Black and White), 249.6 vs 251.7 ms (Color, 6 colours), 214.1 vs
218.6 ms (Color, 6 colours, Noise 1), same segment counts.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_018tBcdwNhptqtmX32JCGFzB
The UI parity row (and G12 in gaps.md, M12 in roadmap.md) said quick Image
Trace presets reveal the editable controls. Since baadbe2c6 a quick trace
leaves the panel as it was (test a_quick_trace_leaves_the_panel_as_it_was)
and the Image Trace button in the Control bar or Properties opens the
controls. That matches the public documentation: choosing a preset traces
the image with it straight away, and the panel is opened with the Image
Trace button in the Control panel or the Properties panel's Image Trace
section
(https://helpx.adobe.com/illustrator/using/image-trace-results-optimization.html).
Reworded the three rows, cited the URL, added revision-history entries and
status-line notes. Wording only; scores, statuses and hours unchanged.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_018tBcdwNhptqtmX32JCGFzB
Abutting traces must draw each edge between two neighbouring regions once:
the filled paths on either side follow one curve, one of them reversed, and
together cover the image without gaps or overlaps. These tests check that
from the trace output alone, with no renderer: points along each path's
outline must lie within 0.001 px of another path's outline, the paths' areas
must sum to the image's, and every point of a 4 x 4 per-pixel grid must lie
in exactly one path.
The images are drawn in code, hard-edged and anti-aliased, and traced in
Color (Abutting) and Black and White, each at the default settings, Paths and
Corners 0 and 100, Snap Curves To Lines, and Shapes. They cover a closed island
in a background, a ring with and without a core of a third colour, a wavy
two-colour seam, a three-region junction, outlines that leave a junction and
return to it (the 111 / 101 / 110 pattern at 1, 2, 3 and 6 px, and framed),
squares touching at a corner, a one-pixel hole and a one-pixel island (kept at
Noise 1, dropped together with what they leave at Noise 25), separate areas of
one colour, and slanted hatches with one-pixel paper channels. A further test
shows the check finds a 0.01 px shift and stacked (Overlapping) layers.
Two tests are ignored as documented current gaps:
- At Paths 0 and Corners 0, a piece returning to its own junction collapses:
its region is dropped and its area left uncovered (draft #2205 fixes it).
- With Shapes, a region whose outline runs through a junction becomes an exact
shape its neighbour does not share (up to 0.0039 px in Black and White).
No tracer code changes.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_018tBcdwNhptqtmX32JCGFzB
In Grayscale mode with Gradients on, the trace could fill paths with
coloured linear gradients (for example a blue-to-orange ramp) beside
grey flat fills. Gradient detection ran on the RGB image, so the stops
and the gradient's palette colour (its middle) were the source colours,
while Grayscale's flat fills come from each pixel's luma.
Grayscale now finds gradients, and settles their edges, in a copy of the
image reduced to the same Rec. 601 luma the flat grays are quantised
from. Every channel of that copy is equal, so the fitted stops, the ends
carried to the area's edges and the palette entry are all grey. Color
mode is unchanged: it still detects on the RGB image (trace output for
gradient fixtures is byte-identical before and after).
Adds grayscale_gradients_are_grey: a code-drawn sky ramp over flat
shapes with a ramp rectangle, traced in Grayscale with 4 grays and
Gradients on, must still produce a gradient, and every path colour,
gradient stop and palette entry must be grey.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_018tBcdwNhptqtmX32JCGFzB
When a long horizontal or vertical run crosses a three-colour junction one
or two pixels before a corner, the corner correction for that run borrowed
a neighbouring crack's shift that the region on the other side of the
junction never applied, leaving hairline gaps between abutting colours.
The correction now applies only when all three cracks it uses lie in the
same shared piece.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_018tBcdwNhptqtmX32JCGFzB
The 'Last updated · Change' status lines in docs/roadmap.md, docs/target-app-parity.md and docs/image-trace-quality.md change with almost every merge to main, so this PR's edits to them conflicted each time main moved. They are reverted to main's text; this PR's evidence bullets, history rows and table entries stay.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_018tBcdwNhptqtmX32JCGFzB
The 'Last updated · Change' status lines in docs/roadmap.md, docs/target-app-parity.md and docs/image-trace-quality.md change with almost every merge to main, so this PR's edits to them conflicted each time main moved. They are reverted to main's text; this PR's evidence bullets, history rows and table entries stay.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_018tBcdwNhptqtmX32JCGFzB
In Grayscale mode with Gradients on, the trace could fill paths with
coloured linear gradients (for example a blue-to-orange ramp) beside
grey flat fills. Gradient detection ran on the RGB image, so the stops
and the gradient's palette colour (its middle) were the source colours,
while Grayscale's flat fills come from each pixel's luma.
Grayscale now finds gradients, and settles their edges, in a copy of the
image reduced to the same Rec. 601 luma the flat grays are quantised
from. Every channel of that copy is equal, so the fitted stops, the ends
carried to the area's edges and the palette entry are all grey. Color
mode is unchanged: it still detects on the RGB image (trace output for
gradient fixtures is byte-identical before and after).
Adds grayscale_gradients_are_grey: a code-drawn sky ramp over flat
shapes with a ramp rectangle, traced in Grayscale with 4 grays and
Gradients on, must still produce a gradient, and every path colour,
gradient stop and palette entry must be grey.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_018tBcdwNhptqtmX32JCGFzB
Abutting traces must draw each edge between two neighbouring regions once:
the filled paths on either side follow one curve, one of them reversed, and
together cover the image without gaps or overlaps. These tests check that
from the trace output alone, with no renderer: points along each path's
outline must lie within 0.001 px of another path's outline, the paths' areas
must sum to the image's, and every point of a 4 x 4 per-pixel grid must lie
in exactly one path.
The images are drawn in code, hard-edged and anti-aliased, and traced in
Color (Abutting) and Black and White, each at the default settings, Paths and
Corners 0 and 100, Snap Curves To Lines, and Shapes. They cover a closed island
in a background, a ring with and without a core of a third colour, a wavy
two-colour seam, a three-region junction, outlines that leave a junction and
return to it (the 111 / 101 / 110 pattern at 1, 2, 3 and 6 px, and framed),
squares touching at a corner, a one-pixel hole and a one-pixel island (kept at
Noise 1, dropped together with what they leave at Noise 25), separate areas of
one colour, and slanted hatches with one-pixel paper channels. A further test
shows the check finds a 0.01 px shift and stacked (Overlapping) layers.
Two tests are ignored as documented current gaps:
- At Paths 0 and Corners 0, a piece returning to its own junction collapses:
its region is dropped and its area left uncovered (draft #2205 fixes it).
- With Shapes, a region whose outline runs through a junction becomes an exact
shape its neighbour does not share (up to 0.0039 px in Black and White).
No tracer code changes.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_018tBcdwNhptqtmX32JCGFzB
When a long horizontal or vertical run crosses a three-colour junction one
or two pixels before a corner, the corner correction for that run borrowed
a neighbouring crack's shift that the region on the other side of the
junction never applied, leaving hairline gaps between abutting colours.
The correction now applies only when all three cracks it uses lie in the
same shared piece.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_018tBcdwNhptqtmX32JCGFzB
