Retrainable-RTMO-s / README.md
Luigi's picture
Update README.md
e2f80db verified
metadata
license: apache-2.0
datasets:
  - phiyodr/coco2017
pipeline_tag: keypoint-detection
tags:
  - 2D_Pose_Estimation
  - MMPOSE
  - RTMO

Retrainable RTMO-s Model

This repository provides a fully retrainable RTMO-s checkpoint for 2D human pose estimation in the MMPOSE framework.

The RTMO one-stage model family (variants T, S, M, L) was originally released by the OpenMMLab team with pre-trained weights, but those official PyTorch .pth checkpoints do not preserve all parameter keys—making fine-tuning or re-training impossible within MMPOSE (see discussion #3076).

To address this limitation, PESI has faithfully reproduced the RTMO-s training procedure on the MS COCO 2017 dataset using the exact configuration from the official MMPOSE RTMO project. Our checkpoint preserves every model key, enabling you to fine-tune on custom datasets (e.g. Body7, MPII) or continue training from this strong baseline.

Key Features

  • One-stage RTMO-s architecture as described in the MMPOSE RTMO project
  • All keys retained in the PyTorch checkpoint for full retrainability (unlike the official weights)

References

  1. OpenMMLab MMPOSE RTMO project: “RTMO: Real-time One-Stage Multi-person Pose Estimation”
  2. Issue discussion on missing keys in official RTMO checkpoints