Skip to content

Latest commit

 

History

History
 
 

pose_estimation

README.md

COCO pose estimation with UniFormer

We currenent release the code and models for:

  • Top-down

Updates

01/18/2022

[Initial commits]:

  1. Models with Top-down.

Model Zoo

The followed models and logs can be downloaded on Google Drive: total_models, total_logs.

We also release the models on Baidu Cloud: total_models (lqht), total_logs (j7e2).

Note

Top-Down

Backbone Input Size AP AP50 AP75 ARM ARL AR FLOPs Model Log Shell
UniFormer-S 256x192 74.0 90.3 82.2 66.8 76.7 79.5 4.7G google google run.sh/config
UniFormer-S 384x288 75.9 90.6 83.4 68.6 79.0 81.4 11.1G google google run.sh/config
UniFormer-S 448x320 76.2 90.6 83.2 68.6 79.4 81.4 14.8G google google run.sh/config
UniFormer-B 256x192 75.0 90.6 83.0 67.8 77.7 80.4 9.2G google google run.sh/config
UniFormer-B 384x288 76.7 90.8 84.0 69.3 79.7 81.4 14.8G google google run.sh/config
UniFormer-B 448x320 77.4 91.1 84.4 70.2 80.6 82.5 29.6G google google run.sh/config

Usage

Installation

Please refer to get_started for installation and dataset preparation.

Training

  1. Download the pretrained models in our repository.

  2. Simply run the training scripts in exp as followed:

    bash ./exp/top_down_256x192_global_small/run.sh

    Or you can train other models as follower:

    # single-gpu training
python tools/train.py <CONFIG_FILE> --cfg-options model.backbone.pretrained_path=<PRETRAIN_MODEL> [other optional arguments]

# multi-gpu training
tools/dist_train.sh <CONFIG_FILE> <GPU_NUM> --cfg-options model.backbone.pretrained_path=<PRETRAIN_MODEL> [other optional arguments]

[Note]:

  • We use global MHRA during training and set the corresponding hyperparameters in the config.py:

    window: False, # whether use window MHRA
hybrid: False, # whether use hybrid MHRA

  • To avoid out of memory, we can use torch.utils.checkpoint in the config.py:

    use_checkpoint=True, # whether use checkpoint
checkpoint_num=[0, 0, 2, 0], # index for using checkpoint in every stage

Testing

# single-gpu testing
python tools/test.py <CONFIG_FILE> <POSE_CHECKPOINT_FILE> --eval mAP

# multi-gpu testing
tools/dist_test.sh <CONFIG_FILE> <POSE_CHECKPOINT_FILE> <GPU_NUM> --eval mAP

Acknowledgement

This repository is built based on mmpose and HRT repository.