Simple Baselines for Human Pose Estimation and Tracking
2018-04-17ECCV 2018Code Available1· sign in to hype
Bin Xiao, Haiping Wu, Yichen Wei
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/leoxiaobin/pose.pytorchOfficialIn paperpytorch★ 0
- github.com/PaddlePaddle/PaddleDetectionpaddle★ 14,132
- github.com/open-mmlab/mmposepytorch★ 7,439
- github.com/osmr/imgclsmobmxnet★ 3,015
- github.com/mindspore-lab/mindonemindspore★ 464
- github.com/leeyegy/SimDRpytorch★ 340
- github.com/leeyegy/simccpytorch★ 340
- github.com/simochen/flowtrack.pytorchpytorch★ 0
- github.com/mks0601/TF-SimpleHumanPosetf★ 0
- github.com/Microsoft/human-pose-estimation.pytorchpytorch★ 0
Abstract
There has been significant progress on pose estimation and increasing interests on pose tracking in recent years. At the same time, the overall algorithm and system complexity increases as well, making the algorithm analysis and comparison more difficult. This work provides simple and effective baseline methods. They are helpful for inspiring and evaluating new ideas for the field. State-of-the-art results are achieved on challenging benchmarks. The code will be available at https://github.com/leoxiaobin/pose.pytorch.
Tasks
Benchmark Results
| Dataset | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| JHMDB (2D poses only) | SimplePose | PCK | 94.4 | — | Unverified |
| OCHuman | ResNet-50 | Test AP | 30.4 | — | Unverified |