Ffmpeg keyframe extraction

Skill cxcscmu/SkillLearnBench/skills/b1-one-shot-claude-haiku-4-5/video-object-counting/ffmpeg-keyframe-extraction

[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.

Install

npx -y skills add cxcscmu/SkillLearnBench --skill ffmpeg-keyframe-extraction

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its author says it does

Copied from the file, not written here

Extract key frames (I-frames) from video files using FFmpeg to identify scene changes and important moments.

SKILL.md

2.5 KB, as published. Nobody here has run it

FFmpeg Keyframe Extraction

Overview

FFmpeg can extract keyframes (I-frames) from video files. Keyframes are complete frames that don't depend on other frames for decoding, making them ideal for analysis and detection tasks.

Installation

# Ubuntu/Debian
sudo apt-get install ffmpeg

# macOS
brew install ffmpeg

Usage

Extract All Keyframes from Video

ffmpeg -i input.mp4 -vf "select=eq(pict_type\,I)" -vsync 0 output_%03d.png

Parameters:

-i input.mp4: Input video file
-vf "select=eq(pict_type\,I)": Filter to select only I-frames (keyframes)
-vsync 0: Output one image per keyframe (important for proper numbering)
output_%03d.png: Output pattern with zero-padded frame numbers (001, 002, etc.)

Extract Keyframes with Frame Rate Control

ffmpeg -i input.mp4 -vf "fps=1/5,select=eq(pict_type\,I)" -vsync 0 output_%03d.png

This extracts at most 1 keyframe every 5 seconds.

Check Video Information

ffprobe -v error -show_entries format=duration -of default=noprint_wrappers=1:nokey=1:noprint_wrappers=1 video.mp4

Output Naming Convention

When using output_%03d.png:

Frame 1: output_001.png
Frame 2: output_002.png
Frame 100: output_100.png

Python Integration

import subprocess
import os

def extract_keyframes(video_path, output_dir, output_prefix="keyframes"):
    """Extract keyframes from video file"""
    os.makedirs(output_dir, exist_ok=True)
    output_pattern = os.path.join(output_dir, f"{output_prefix}_%03d.png")

    cmd = [
        'ffmpeg',
        '-i', video_path,
        '-vf', 'select=eq(pict_type\\,I)',
        '-vsync', '0',
        output_pattern
    ]

    result = subprocess.run(cmd, capture_output=True, text=True)
    if result.returncode != 0:
        raise RuntimeError(f"FFmpeg error: {result.stderr}")

    # Return list of extracted frames
    frames = sorted([
        os.path.join(output_dir, f)
        for f in os.listdir(output_dir)
        if f.startswith(output_prefix)
    ])
    return frames

Tips

Keyframes are typically smaller in file size than regular frames
For consistent results, always use -vsync 0 with keyframe extraction
Output pattern with %03d ensures consistent naming for sorting
Check that files are being created in the expected output directory