ext/hyperframes
2
0
Fork 0
mirror of https://github.com/heygen-com/hyperframes.git synced 2026-09-03 04:38:33 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix(producer): sample PSNR checkpoints from common duration of rendered+snapshot

Browse Source 
Four regression tests (font-variant-numeric, many-cuts, missing-host-comp-id,
variables-prod) failed on this PR with `Unable to parse PSNR output at <last
checkpoint>s`. Root cause: the harness derived all 100 checkpoints from the
*rendered* video's container duration, then asked ffmpeg's PSNR filter to
compare the same frame index from both videos.

The encoder changes earlier in this PR add `-avoid_negative_ts make_zero` to
the mux step. With AAC audio that shifts the first audio sample to t=0
instead of the encoder-delay offset, extending reported container duration
by ~20ms without changing video frame count. For the four failing tests,
the i=99 checkpoint then landed on a frame index that exists in the rendered
video but not in the snapshot baseline (e.g. round(2.98998 * 24) = 72 in a
72-frame baseline). ffmpeg's PSNR filter ran on zero matched frames and
emitted no `average:` line, so the parser threw.

Fix: probe both videos and use min(rendered, snapshot) duration when
spreading checkpoints. This is the correct semantics for symmetric PSNR
comparison anyway — both videos must have a frame at every sampled time.
The change is local to the harness; no encoder behavior changes, no
baselines regenerated.

Other regression tests with audio (chat, sub-composition-video,
vignelli-stacking) passed because their checkpoint-99 frame index landed
inside the baseline's frame range with several frames of slack. The four
failing tests had round-number durations where a 20ms drift was enough to
push the last checkpoint past `nb_frames - 1`.
This commit is contained in:
James
2026-05-05 04:18:32 +00:00
parent b8b82fa8d2
commit 7d1d8ead60
1 changed files with 8 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
packages/producer/src/regression-harness.ts
+8 -1
View File
@@ -650,7 +650,14 @@ async function runTestSuite(
// Visual comparison (100 frames, 1 per 1% of video duration)
logPretty("Comparing visual quality (100 checkpoints)...", "🔍");
const videoMetadata = await extractMediaMetadata(renderedOutputPath);
const videoDuration = videoMetadata.durationSeconds;
const snapshotMetadata = await extractMediaMetadata(snapshotVideoPath);
// Sample at the common duration. Container duration can drift between
// rendered and snapshot when encoder/mux flags change (e.g. -avoid_negative_ts
// can shift the first audio sample, extending reported duration without
// changing video frame count). Using the rendered duration alone makes the
// last checkpoint land on a frame index that may not exist in the snapshot,
// which causes ffmpeg's PSNR filter to emit no `average:` line.
const videoDuration = Math.min(videoMetadata.durationSeconds, snapshotMetadata.durationSeconds);
const visualCheckpoints: Array<{ time: number; psnr: number; passed: boolean }> = [];
for (let i = 0; i < 100; i++) {