Mirror of CMU OpenPose pretrained models (BODY_25, face, hand)

Read full text
Authors: Brian Herreño, CMU Perceptual Computing Lab (como creator principal)
Cited by: 0

Abstract

This dataset is an unmodified mirror of the pretrained Caffe model files required by CMU OpenPose (BODY_25, face, and hand pose models). These model files are part of the OpenPose project and are not authored or modified by the uploader. The original model download servers referenced by the OpenPose installation scripts are currently unreliable or unavailable, which hinders reproducible installation of OpenPose for research and engineering workflows. To address this, the model files have been made available here strictly for preservation and reproducibility purposes, enabling stable and repeatable installations of OpenPose-based pipelines. All rights remain with the original OpenPose authors (CMU Perceptual Computing Lab). No modifications have been made to the model files, and this mirror is provided only to ensure long-term availability. For more context on the model availability situation and the community discussion around it, see the following comment in the OpenPose issue tracker:https://github.com/CMU-Perceptual-Computing-Lab/openpose/issues/2233#issuecomment-2592322909

Fields of Study & Subjects

Human Pose and Action Recognition Hand Gesture Recognition Systems Face recognition and analysis

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
18 January 2026
Document type
Dataset