videobeaux/docs/programs/utilities/num_edits.md

num_edits

Description

Analyzes a timeline or cut structure to count edits, transitions, or shot boundaries for editorial statistics, QC, or structural analysis.

Purpose

num_edits provides fast editorial metrics by identifying the number of cuts, transitions, or shot boundaries in a video.
This helps with:

• QC validation,
• editorial pacing analysis,
• archive metadata generation,
• automated editing metrics,
• machine-learning dataset preparation.

How It Works

1. Shot Boundary Detection
   Uses visual or luminance-based thresholds to detect edits between shots.
2. Event Counting
   Each boundary event increments the edit count.
3. Reporting
   The final count is embedded or returned based on videobeaux pipeline logic.
4. Speed-Optimized
   Designed for rapid scanning across long-form footage or batch libraries.

Program Template

videobeaux -P num_edits \
  -i input.mp4 \
  -o output.mp4 \
  --count VALUE

Arguments

• count — Enables or configures the edit-counting logic. Usually true or a mode such as basic vs detailed depending on implementation.

Real World Example

videobeaux -P num_edits \
  -i myvideo.mp4 \
  -o num_edits_styled.mp4 \
  --count true

Technical Notes

• Detection may use pixel-wise difference, histogram deltas, or scene-change detection filters under the hood.
• Thresholds vary with footage type; high-motion sequences may generate more detected edits.
• The tool does not modify video; it only analyzes structure.
• Useful in workflows that require edit-density analytics or QC verification.

Recommended Usage

• Counting edits in commercials, music videos, or high-cut-rate content.
• Generating metadata for cataloging systems or research datasets.
• Automated QC workflows to detect unexpected edit patterns.
• Comparing pacing between cuts or across versions of an edit.

Quality Tips

• Use higher thresholds for shaky or handheld footage to avoid false positives.
• Use lower thresholds when analyzing animation or motion-poor material.
• For statistical studies, run the tool consistently with identical settings across all sources.