idealo/imagededup

Health v2 · maintenance only

Adoption not part of the score

5666 stars · 476 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

imagededup is a Python library for finding exact and near-duplicate images in a collection using perceptual hashing algorithms (PHash, DHash, WHash, AHash) and CNN-based embeddings. It also provides an evaluation framework to judge deduplication quality against ground truth and tools to visualize found duplicates.

Use cases

• find duplicate images in a folder
• deduplicate an image dataset before training
• detect near-duplicate photos with CNN embeddings
• remove duplicate product images in e-commerce catalogs
• evaluate deduplication quality against ground truth
• generate perceptual hashes for images

When to choose

• you need a simple, well-documented Python API for image deduplication
• you want both hashing-based exact matching and CNN-based near-duplicate detection
• you need to benchmark deduplication results against labeled ground truth

When to avoid

• you need video or audio deduplication
• you require distributed processing of massive image collections beyond a single machine
• you need a ready-made GUI application rather than a library

Facets

library · maturity stable

image-processing computer-vision machine-learning computer-vision image-processing machine-learning python windows cross-platform image-deduplication perceptual-hashing cnn duplicate-detection pytorch linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="idealo/imagededup")