Video to grayscale frames

Skill cxcscmu/SkillLearnBench/skills/b4-skill-creator-claude-haiku-4-5/video-object-counting/video-to-grayscale-frames

[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 video-to-grayscale-frames

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 keyframes from MP4 video files and convert them to grayscale. Use this skill when you need to process video files for analysis by extracting I-frames (keyframes) and converting them to grayscale images for computer vision tasks.

SKILL.md

2.4 KB, as published. Nobody here has run it

Video to Grayscale Frames

This skill guides you through extracting keyframes from MP4 video files and converting them to grayscale images for analysis.

Overview

The workflow has two main steps:

Extract keyframes from the MP4 video using FFmpeg
Convert to grayscale and save frames as PNG files

Step 1: Extract Keyframes from MP4

Use the ffmpeg skill to extract I-frames (keyframes) from the video.

Command Structure

ffmpeg -i <input.mp4> -vf "select=eq(pict_type\,I)" -vsync vfr <output_pattern>

Parameters

input.mp4: Path to your video file
output_pattern: Output file pattern (e.g., /root/keyframes_%03d.png)
- %03d creates zero-padded numbers: keyframes_001.png, keyframes_002.png, etc.

Example

For a video at /root/super-mario.mp4, extracting to /root/keyframes_%03d.png:

ffmpeg -i /root/super-mario.mp4 -vf "select=eq(pict_type\,I)" -vsync vfr /root/keyframes_%03d.png

This creates: keyframes_001.png, keyframes_002.png, keyframes_003.png, etc.

Step 2: Convert Extracted Frames to Grayscale

Use the image_editing skill to convert each extracted keyframe to grayscale IN-PLACE (overwriting the original RGB image).

Conversion Process

For each keyframe image:

Input: /root/keyframes_001.png (RGB)
Operation: Convert to grayscale
Output: /root/keyframes_001.png (grayscale, overwrites original)

Batch Processing

When converting multiple frames, process them sequentially or in parallel depending on available resources. Ensure each frame is converted and saved before proceeding to analysis.

Output Format

After this skill completes:

Keyframes are stored as /root/keyframes_001.png, /root/keyframes_002.png, etc.
All files are in grayscale format
Files are ready for object detection and counting

Common Issues

FFmpeg not installed: The ffmpeg skill should handle installation
No keyframes extracted: Video may have no I-frames; consider extracting all frames with -fps 1 instead
Image format errors: Ensure images are valid PNG files before grayscale conversion