lllyasviel / lllyasviel/ControlNet

Cropping vs. Letterboxing for training sets

Open
#306 4 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
34.1k
Forks
3k
PR merge metrics
No merged PRs in 30d

Description

How are other people dealing with training sets that are landscape images?

- Anyone have luck with just letterboxing? Does transparent letterboxing work?
- Center crop - easy but losing a lot of information...
- Crop left + crop right + crop center - not sure if this would generate too much redundant training images.
- Smart crop - anyone know of a good library for non-face/non-people images?

Just all around curious what others have had success with.

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 discusses landscape training images and compares letterboxing, transparent padding, center and multi-position crops, and smart cropping. No files, tests, implementation entry point, or completion criterion are provided; first seek clarification on whether a concrete preprocessing feature is wanted and how success should be evaluated.

Written by the indexing model from the issue text.

Assessment

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

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.