nextcloud / nextcloud/recognize

Allow for manual face tag

Open
#1,372 5 comments 12 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

enhancement
Dominant language
PHP
Stars
699
Forks
68
Avg merge
1d 4h
Merged PRs (30d)
5

Description

Describe the feature you'd like to request

The ability to manually tag a face in a photo.

Sometimes recognize does not pick up someone in the photo.

Reasons why the face was not found:

  • Either the face has not been found at all,
  • The person is on an angle, or their whole face is not in the frame.
  • The person is wearing a silly mask in a 'selfie booth'

You still want to be able to find the photo of the person when you click on that avatar.

Describe the solution you'd like

The ability to manually tag a face in a photo:
Allowing the system either add the vector to an existing list of vectors for the person.
or create a new record of a person, generate a vector for the tagged location, and assign the vector to the name record.
(Possibly optional tick box of "Add face to index")

If recognize is not "seeing" the person, allowing users to tag a person's face would put in 'rotation' for the next scan.

Describe alternatives you've considered

I have not seen an option apart from lowering tolerances in face detection, then getting swamped

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 tracing the existing face-recognition and face-indexing flow described in the issue. Done means a user can manually tag a face, associate it with an existing or new person record, and find the photo through that person's avatar.

Written by the indexing model from the issue text.

Assessment

Tech stack
machine-learning, php
Domain
computer-vision
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
38/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.