The two rounds were meant to answer the stroke sooner, but a preview cannot be
half a repair: the blurred fill is what a machine without the runtime gets, and
with it on the stage a spot of dirt came back smeared rather than gone — worst of
all on a run of strokes, where each new one cancels the model's round that was
still on its way and leaves the blur standing longest. The repair that works is
the one that waits for the model, and the wait is worth it.
The worker stays available but is switched off with the rest: this reverts the
runtime's proxy along with the second round, so the stage is exactly what it was
before, lag and all.
Reverts 5489114.
SINGLE is where a roll is read one frame at a time, and the step from one frame
to the next was a trip to the strip below the picture and back. The stage now
carries the pair itself, at the edges the eye already reads as "there is more
this way", dimmed at the ends of the roll rather than taken away so the pair does
not jump about. The press stops at the arrow: the zoom, the pan and the open
under it are the picture's.
The phone's full-screen view has no room for a pair, so the same step is the
reader's own finger. A drag that travelled sideways is a step the way the list is
read; one that wandered up and down, and a tap in place, are left alone.
Both walk the list on screen — filters and sort included — so what is drawn after
the step is what stood beside it.
Verified against a local preview with a seeded shelf: the arrows appear, the left
one is dimmed on the newest frame and the right one on the oldest, and a left
swipe on the phone moves to the next frame and back.
The graph is seconds of solid arithmetic however large the crop is, and it ran
where the pointer lives: the brush lagged behind the hand for exactly as long as
the model took. The runtime's wasm provider now works in a worker of its own, so
the page keeps drawing while the repair is thought through.
The wait itself stays — it is the model's, not the page's — so a preview paints
the blurred fill first and lets the model's version of the same mask replace it.
Filing and export still wait for the model: the faster answer is for the eye
waiting on the brush and for nothing else.
Measured on the same photo and stroke, worst main-thread gap 3703 ms before and
487 ms after; previews land at 1.0 s and 5.3 s, and the model's render is the one
left on the stage.
The dust brush filled its hole with a blur of the hole's own rim — the light
beside it, carried in. Right for a speck on a wall, and visibly thin over
anything with structure: a stroke across a shirt came back as the average of a
shirt. The fill is now a learned one.
engine/inpaint.ts runs MI-GAN (andraniksargsyan/migan, the same
`migan_pipeline_v2.onnx` inpaint-web ships) through onnxruntime-web on the
visitor's own machine: 27MB of weights in public/models, WebGPU where the graph
fits the provider, the processor otherwise, one warm-up pass so the first repair
is not the one that pays for the kernels. It is handed the masked crop and
nothing else — the model's mask polarity is 255 KEEP / 0 HOLE, the opposite of
the obvious guess, and the other way round repaints the frame and leaves the
hole (measured).
Step 6e of the renderer cuts the crop at the resolution the model will actually
run at (512 on its longest side — the graph is fully convolutional and would
take a whole frame, and ten seconds over one), builds the mask from the brush's
own circles, and draws the answer into the fill surface at its own place in the
box. The composite, the feathered mask and the untouched frame around it are
unchanged, so nothing outside the mask moves — and when the model cannot be had
the blurred fill is still built first and still stands. The preview gets this
for free: renderPhoto was already async and the preview was already debounced.
Node probe: the model + polarity + read/write path run; heal-check.mjs keeps
measuring the fallback and now asserts the wiring and the polarity. Browser:
engine/inpaint.ts in a real Chromium returns a patch (hole moves, kept pixels
within 0.02 of a level), and the built app renders a 42-spot stroke in 4.7s on
single-threaded wasm with no fallback warning — 1.15s and the warning when the
model is taken away.
3dtours