Lightning-AI / Lightning-AI/torchmetrics

Add OKS metric

Open
#1,361 2 comments 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

enhancement New metric
Dominant language
Python
Stars
2.5k
Forks
526
Avg merge
6d 11h
Merged PRs (30d)
5

Description

## 🚀 Feature

Add support for keypoints inside [torchmetrics map](https://torchmetrics.readthedocs.io/en/stable/detection/mean_average_precision.html#).

### Additional context

Should be fairly simple to do that using the [iou_type](https://github.com/Lightning-AI/metrics/blob/master/src/torchmetrics/detection/mean_ap.py#L314) flag, since [torchmetrics map](https://torchmetrics.readthedocs.io/en/stable/detection/mean_average_precision.html#) is built upon [pycocotools](https://github.com/cocodataset/cocoapi/blob/master/PythonAPI/pycocotools/cocoeval.py#L110).

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 with src/torchmetrics/detection/mean_ap.py, especially the iou_type handling referenced in the issue, and compare it with pycocotools/cocoeval.py. Confirm the documented mean average precision integration and define done as torchmetrics map supporting keypoint evaluation through an appropriate keypoint setting.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
computer-vision, 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.