You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Commit ca6ea3a
Browse filesBrowse the repository at this point in the historyBrowse files
refactor(embedded-captions): fetch PP-MattingV2 model on demand instead of shipping in-tree
The 34 MB ppmattingv2 ONNX was committed as a raw blob (added before the
*.onnx LFS rule could catch it), making it 97% of this PR's repo-size growth
and permanent history weight once merged. Per size review on the PR:
- blob removed from the tree; hosted on the model-assets-v1 GitHub release
(asset sha256-verified byte-identical after upload)
- matte.cjs resolves: MATTE_MODEL env -> legacy bundled copy if present ->
~/.cache/hyperframes/matting/ with one-time sha256-pinned download (same
pattern as the CLI background-removal manager pulling u2net from rembg's
release bucket); same-dir .part temp + atomic rename
- new `matte.cjs --ensure-model` pre-warm flag; SKILL.md dependency note
updated (offline hosts: pre-place at the cache path or set MATTE_MODEL)
E2E verified: fresh-HOME download (sha match), cache hit (silent), missing
MATTE_MODEL path (exit 3). Author-time fetch only — render path untouched.
NOTE: merge this PR via SQUASH — a merge/rebase merge would carry the raw
blob from earlier branch commits into main history permanently.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
Copy file name to clipboardExpand all lines: skills/embedded-captions/SKILL.md
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -294,6 +294,6 @@ track has its own, much simpler spec → **[references/rail.md](references/rail.
294
294
-**Transcription = hyperframes' Whisper** (whisper.cpp — native C++, no API key, no Python). `transcribe.cjs` wraps `hyperframes transcribe`; hyperframes auto-installs whisper.cpp (Homebrew or source build) and the model on first run. Falls back to an existing word-level `transcript.json` if present.
295
295
-**Source video** — `matte.cjs` / `transcribe.cjs` auto-resolve `source.mp4` (or glob the clip / read `hyperframes.json`), so `hyperframes init --video X.mp4` needs no manual rename.
296
296
-**fps** — `matte.cjs` extracts at the source's native rate and records `matte.fps`; `render-and-composite.sh` uses that so the matte stays frame-aligned.
297
-
-`assets/ppmattingv2_stdc1_human_544x960.onnx` (34 MB, Apache-2.0) — **ships with the skill**(no auto-download; the upstream ONNX exports have no stable public URL). `MATTE_MODEL` env can point at another `<H>×<W>.onnx` export (e.g. 1088x1920 for max detail); `MATTE_EMA` tunes temporal smoothing (default 0.5).
297
+
-**PP-MattingV2 model** (`ppmattingv2_stdc1_human_544x960.onnx`, 34 MB, Apache-2.0) — **downloaded on demand**by `matte.cjs` (one-time, sha256-pinned) from the repo's `model-assets-v1` GitHub release into `~/.cache/hyperframes/matting/` (the upstream ONNX exports have no stable public URL, so we host the exact blob ourselves). Pre-warm with `node scripts/matte.cjs --ensure-model`; offline hosts: pre-place the file at the cache path or set `MATTE_MODEL`. `MATTE_MODEL` env can point at another `<H>×<W>.onnx` export (e.g. 1088x1920 for max detail); `MATTE_EMA` tunes temporal smoothing (default 0.5).
298
298
299
299
If a hard dependency is missing, STOP and ask the user — don't silently skip steps.
0 commit comments