tensorflow / tensorflow/datasets

Automatically rotate images based on image metadata

Open
#3,202 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

enhancement
Dominant language
Python
Stars
4.6k
Forks
1.6k
Avg merge
3h 54m
Merged PRs (30d)
1

Description

Is your feature request related to a problem? Please describe.
It seems some datasets may have a few images which aren't oriented the right way, while the metadata of underlying image file gives the orientation.

eg: open_images_v4, 0000587b9647679e.jpg

Describe the solution you'd like
Image should automatically be rotated by TFDS at preparation time.
Bounding box if any should also be rotated? (or optionally so)

Describe alternatives you've considered
None

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 locating the TFDS image preparation path and the Open Images v4 example named in the issue. Determine how image orientation metadata should affect images and whether bounding boxes must rotate as well; done means prepared images have the intended orientation with a defined bounding-box behavior.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, tensorflow
Domain
computer-vision, data, machine-learning
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.