ageitgey / ageitgey/face_recognition

Face Encoding with smaller length (less than 128)

Open
#882 0 comments 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
56.8k
Forks
13.7k
PR merge metrics
No merged PRs in 30d

Description

  • face_recognition version: 1.2.3
  • Python version: 3.6.8
  • Operating System: Ubuntu 16.04
Description

I wanted to know if it's possible to get face encoding with smaller model with less than 128 components.
For example if we are searching among over 1M faces, it may help with first finding the top 100 nearest ones based on smaller model with 32 components (features), and then on those 100 faces find the nearest one using the 128 component face encoding or even a larger and more accurate model.

Contributor guide

Open the contributing guide

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

Start by reviewing the face encoding API and its current 128-component output; the issue names no implementation file or test. Determine whether lower-dimensional encodings are supported and whether the proposed two-stage nearest-neighbor workflow belongs in the library. Done means a clear supported behavior or limitation is documented and validated.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
computer-vision, machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.