SimplePose
Code and pre-trained models for our paper, “Simple Pose: Rethinking and Improving a Bottom-up Approach for Multi-Person Pose Estimation”, accepted by AAAI-2020.
Also this repo serves as the Part B of our paper "Multi-Person Pose Estimation Based on Gaussian Response Heatmaps" (under review). The Part A is available at this link.
-
Update
A faster project is to be released.
Introduction
A bottom-up approach for the problem of multi-person pose estimation.
Contents
- Training
- Evaluation
- Demo
Project Features
- Implement the models using Pytorch in auto mixed-precision (using Nvidia Apex).
- Support training on multiple GPUs (over 90% GPU usage rate on each GPU card).
- Fast data preparing and augmentation during training (generating about 40 samples per second on signle CPU process and much more if wrapped by DataLoader Class).
- Focal L2 loss.
- Multi-scale supervision.
- This project can also serve as a detailed practice to the green hand in Pytorch.
Prepare
-
Install packages:
Python=3.6, Pytorch>1.0, Nvidia Apex and other packages needed.
-
Download the COCO dataset.
-
Download the pre-trained models (default configuration: download the pretrained model snapshotted at epoch 52 provided as follow).
Download Link: BaiduCloud
Alternatively, download the pre-trained model without optimizer checkpoint only for the default configuration via GoogleDrive
-
Change the paths in the code according to your environment.
Run a Demo
python demo_image.py
Inference Speed
The speed of our system is tested on the MS-COCO test-dev dataset.
- Inference speed of our 4-stage IMHN with 512 × 512 input on one 2080TI GPU: 38.5 FPS (100% GPU-Util).
- Processing speed of the keypoint assignment algorithm part that is implemented in pure Python and a single process on Intel Xeon E5-2620 CPU: 5.2 FPS (has not been well accelerated).
Evaluation Steps
The corresponding code is in pure python without multiprocess for now.
python evaluate.py
Results on MSCOCO 2017 test-dev subset (focal L2 loss with gamma=2):
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets= 20 ] = 0.685
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets= 20 ] = 0.867
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets= 20 ] = 0.749
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets= 20 ] = 0.664
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets= 20 ] = 0.719
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 20 ] = 0.728
Average Recall (AR) @[ IoU=0.50 | area= all | maxDets= 20 ] = 0.892
Average Recall (AR) @[ IoU=0.75 | area= all | maxDets= 20 ] = 0.782
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets= 20 ] = 0.688
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets= 20 ] = 0.784
Training Steps
Before training, prepare the training data using ''SimplePose/data/coco_masks_hdf5.py''.
Multiple GUPs are recommended to use to speed up the training process, but we support different training options.
-
Most code has been provided already, you can train the model with.
- 'train.py': single training process on one GPU only.
- 'train_parallel.py': signle training process on multiple GPUs using Dataparallel.
- 'train_distributed.py' (recommended): multiple training processes on multiple GPUs using Distributed Training:
python -m torch.distributed.launch --nproc_per_node=4 train_distributed.py
Note: The loss_model_parrel.py is for train.py and train_parallel.py, while the loss_model.py is for train_distributed.py and train_distributed_SWA.py. They are different in dividing the batch size. Please refer to the code about the different choices.
For distributed training, the real batch_size = batch_size_in_config* × GPU_Num (world_size actually). For others, the real batch_size = batch_size_in_config*. The differences come from the different mechanisms of data parallel training and distributed training.
Referred Repositories (mainly)
- Realtime Multi-Person Pose Estimation version 1
- Realtime Multi-Person Pose Estimation version 2
- Realtime Multi-Person Pose Estimation version 3
- Associative Embedding
- NVIDIA/apex
Recommend Repositories
Faster Version: Chun-Ming Su has rebuilt and improved the post-processing speed of this repo using C++, and the improved system can run up to 7~8 FPS using a single scale with flipping on a 2080 TI GPU. Many thanks to Chun-Ming Su.
Citation
Please kindly cite this paper in your publications if it helps your research.
@inproceedings{li2020simple,
title={Simple Pose: Rethinking and Improving a Bottom-up Approach for Multi-Person Pose Estimation.},
author={Li, Jia and Su, Wen and Wang, Zengfu},
booktitle={AAAI},
pages={11354--11361},
year={2020}
}
677 Dec 25, 2022
108 Dec 1, 2022
367 Dec 27, 2022
66 Jan 4, 2023
160 Dec 21, 2022
157 Nov 12, 2022
4.9k Dec 31, 2022
79 Dec 30, 2022
174 Dec 19, 2022
25.7k Jan 9, 2023
49 Dec 22, 2022
173 Dec 14, 2022
26 May 26, 2022
42 Nov 24, 2022
4 Nov 3, 2022
5.8k Dec 31, 2022
40 Dec 30, 2022
8 Sep 14, 2022
328 Dec 17, 2022