LAION-AI / LAION-AI/laion-dedup

Dedup in LAION-A

Open
#4 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
18
Forks
2
PR merge metrics
No merged PRs in 30d

Description

Hi Romain, I have found millions of duplicated images from the clusters of the laion5b_index. I noticed that you have already dedup LAION-2B-en to LAION-A by pHash: "LAION-A is a 900M subset of LAION-2B filtered with aesthetic V2 4.5+ and pHash deduplicated [LAION-A] Would you tell me more about how you do dedup for LAION-A?

Currently I am guessing using pHash to generate bit vectors, use faiss to build a binary IVF index, and then select the clusters in the IVF index with group similarity > threshold for deduplication. Is this the process underlying LAION-A?

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

The issue names LAION-A, LAION-2B, laion5b_index, pHash, and a proposed FAISS binary IVF approach, but it mentions no repository files, tests, or entry points. Start by inspecting the repository's Python entry points and any deduplication documentation, then compare them with the cited LAION-A description; done means documenting or confirming the actual deduplication process.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
computer-vision, data
Issue type
Documentation
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.