human-extractor

Community

Crop human detections from dashcam videos.

Authoryousufjoyian
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill automates the detection and extraction of human subjects from dashcam video footage, enabling rapid retrieval of evidence-ready crops and searchable metadata.

Core Features & Use Cases

  • GPU-accelerated: YOLOv8-based person detection with ByteTrack for stable IDs across frames.
  • Cropped outputs & indexing: saves cropped human images and builds per-date MANIFEST.csv and a global INDEX.csv for fast search.
  • Resume-safe: idempotent processing that skips already processed crops, allowing interruption and resumption.
  • Optional filtering: CLIP-based head-covering classification can be integrated in extended pipelines.
  • Use Case: process Park_R videos to extract all detected humans, generate per-date data, and enable cross-date searches.

Quick Start

Run the extraction workflow by providing a JSON config that points to source dashcam directories and an output folder. Example: create config.json from assets/config_template.json and run:

  • python scripts/run_extraction.py config.json
  • Or adapt the template to your environment and execute.

Dependency Matrix

Required Modules

None required

Components

scriptsreferencesassets

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: human-extractor
Download link: https://github.com/yousufjoyian/claude-skills/archive/main.zip#human-extractor

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 223,000+ vetted skills library on demand.