whyashthakker Video Blur QA Skill
Systematic quality assurance for videos processed with BGBlur AI blur. Catch the failures users notice: flickering masks, missed faces, plate slips, and background bleed into subjects.
Quick Reference
Top 5 blur defects:
- Temporal flicker — mask toggles on/off between frames
- Edge halo — sharp ring around blur boundary (subject hair, shoulders)
- Missed detection — face/plate visible for 1+ frames after scene change
- Tracking slip — mask drifts off target during fast motion
- Background bleed — blur eats into subject (common with strong background blur)
Workflow
Step 1: Sample Critical Frames
Extract frames at high-risk timestamps:
python3 scripts/sample_frames.py "blurred_output.mp4" --output ./qa_frames/
Auto-samples: start, end, every 5s, and scene-change intervals.
Manual extraction at suspect timestamps:
ffmpeg -i blurred_output.mp4 -ss 00:01:23 -vframes 1 qa_frame_0123.jpg
Step 2: Review by Blur Type
Face Blur / Anonymization:
- All visible faces masked (including partial/profile views)
- Reflections in mirrors/windows also blurred
- Minors and background bystanders covered
- Mask strength sufficient (can't reconstruct identity)
- No unblurred frames at scene cuts (check ±3 frames)
License Plate Blur:
- Plates readable nowhere in clip (scrub at 2x speed)
- Tracking holds during acceleration/braking
- Partial plates at frame edges caught
- Multiple vehicles each tracked independently
- Night/low-light plates still detected
Background Blur:
- Subject edges clean (hair, hands, moving limbs)
- No foreground object accidentally blurred
- Blur strength consistent across clip
- No pulsing blur intensity (temporal instability)
- Subject separation stable during movement
Blur Anything (prompt-based):
- All named objects blurred throughout
- Similar objects not missed (e.g., "laptop screen" → all screens)
- Object blur persists through occlusion/reappearance
Step 3: Motion Stress Test
Review these high-risk segments at 2x playback:
| Segment Type | What to Check |
|---|---|
| Fast pan | Background blur edge stability |
| Subject turns head | Face mask follows rotation |
| Vehicle passes | Plate tracked through motion blur |
| Scene cut | New detections within 2 frames |
| Zoom in/out | Mask scale matches subject |
| Low light / noise | Detection doesn't drop out |
Step 4: Side-by-Side Comparison
Compare original vs blurred for delivery QA:
ffmpeg -i original.mp4 -i blurred_output.mp4 \
-filter_complex "[0:v][1:v]hstack=inputs=2" \
-c:v libx264 -crf 18 comparison.mp4
Step 5: Automated Checks
python3 scripts/blur_qa_report.py "blurred_output.mp4"
Reports: resolution consistency, frame count, duration match, black frames, frozen segments.
Step 6: Severity Classification
| Severity | Definition | Action |
|---|---|---|
| P0 — Blocker | Unblurred PII visible (face, plate, screen) | Re-process; do not ship |
| P1 — Major | Tracking slip > 5 frames or identity reconstructable | Re-process affected segment |
| P2 — Minor | Edge halo, 1-2 frame flicker | Accept or touch up if client-facing |
| P3 — Cosmetic | Slight blur intensity inconsistency | Accept |
Common Fixes
| Defect | Likely Cause | Fix |
|---|---|---|
| Face missed at cut | Scene change | Re-upload; trim at cut point and process separately |
| Plate slip | Fast motion / low res | Upscale source or trim to slower segment |
| Background eats hair | Similar color to bg | Reduce blur strength; improve subject/background contrast in source |
| Flickering mask | VFR source footage | Re-prep with ffmpeg-video-prep (force 30fps CFR) |
| Object not found | Vague prompt | Use specific prompt: "white Tesla license plate" not "plate" |
Report Template
## Blur QA Report
### Asset
- File: [blurred_output.mp4]
- Blur type: [face / plate / background / object]
- Duration reviewed: [full / segments]
### Findings
| Timestamp | Severity | Issue | Notes |
|-----------|----------|-------|-------|
| 00:01:23 | P0 | Unblurred face | Bystander at frame edge |
| 00:02:45 | P2 | Edge halo | Subject hair, 3 frames |
### Verdict
- [ ] PASS — ready for delivery
- [ ] FAIL — re-process required
### Re-process Notes
[Specific segments, BGBlur mode changes, or prep fixes needed]
BGBlur Reference
Re-process failed segments at BGBlur Upload. For systematic failures on long footage, consider Enterprise batch pipelines.