skills.rest
erichowens
erichowensCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

photo-content-recognition-curation-expert

Detect near-duplicate photos and cluster faces in large libraries.

181|30|Updated Nov 16, 2025
Search Tags:#indexing#recognition#deduplication#clustering#photo#burst
One-click install
npx skills add https://github.com/erichowens/some_claude_skills --skill photo-content-recognition-curation-expert
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: photo-content-recognition-curation-expert
Source: https://github.com/erichowens/some_claude_skills/tree/main/.claude/skills/photo-content-recognition-curation-expert
Command: npx skills add https://github.com/erichowens/some_claude_skills --skill photo-content-recognition-curation-expert

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill provides expert capabilities for analyzing photo content and automating curation. It combines traditional computer vision with modern deep learning to recognize faces, animals, and scenes, cluster people, detect duplicates, select the best burst frames, and index large photo libraries efficiently. It reduces manual cataloging time and helps build high-quality organized collections.

Core Features & Use Cases

  • Face recognition & clustering (Apple Photos-style approaches)
  • Pet/animal recognition and clustering
  • Near-duplicate detection using perceptual hashing (DINOHash, pHash, dHash)
  • Burst photo selection with multi-criteria scoring (sharpness, faces, aesthetics)
  • Content detection (screenshots, memes, NSFW filtering)
  • Quick indexing for large libraries (10K+ photos) with batch processing
  • Aesthetic quality scoring (e.g., NIMA)
  • Fast indexing pipelines and scalable storage of embeddings and metadata

Quick Start

  • Use for typical photo libraries to identify people, duplicates, or best shots.
  • Example prompts: "Run face recognition on the album and create person clusters." OR "Index 10K photos and detect duplicates."

Frequently Asked Questions about photo-content-recognition-curation-expert

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I detect and remove near-duplicate photos from a large library?▼

Near-duplicate detection uses perceptual hashing (DINOHash, pHash, dHash) to identify visually similar images across tens of thousands of photos, then clusters and removes redundant frames while preserving the best quality shots.

Can I automatically recognize and cluster faces and pets in my photo collection?▼

Face recognition applies MTCNN and InceptionResnetV1 embeddings to detect and cluster people, while pet recognition uses similar content-detection signals to group animals, enabling Apple Photos-style automated organization.

What's the best way to select the sharpest photo from burst sequences?▼

Burst detection and multi-criteria scoring evaluate sharpness, face presence, and aesthetic quality to automatically select the best frame from each burst sequence, reducing manual review time.

How do I index and organize 10,000+ photos efficiently?▼

GPU-accelerated batch processing with caching and incremental updates creates fast indexing pipelines for large libraries, storing embeddings and metadata for quick retrieval and content-based organization.

Does this approach filter screenshots and NSFW content automatically?▼

Content detection signals distinguish screenshots from photos and apply NSFW screening to flag or remove inappropriate images, while CLIP embeddings and color palettes enable semantic and aesthetic filtering.

What techniques does this use to organize photos by quality and content?▼

A hybrid pipeline combines aesthetic quality scoring (NIMA), color palette analysis, and CLIP embeddings to rank and curate photos by visual appeal and semantic content, producing collage-ready outputs.

Related Skills

Explore More Data & Analytics Skills→
xongtaoxongtao

image-ocr

Extract text from PNG, JPEG, TIFF, and BMP images via Tesseract OCR.

Community
Intermediate
mocchaleramocchalera

select-photos

Select event photos into album_select and digest_highlights deliverables.

Community
Intermediate
cas-bigdatalabcas-bigdatalab
541

image_aesthetics_filter

Filter images by aesthetic quality scores using a pre-trained model.

Official
Intermediate
ThedouglerThedougler

photo-content-recognition-curation-expert

Recognize faces, pets, and objects while filtering duplicate photos.

Community
Advanced