---

OpenGait is a flexible and extensible gait recognition project provided by the Shiqi Yu Group and supported in part by WATRIX.AI.

What's New

• [Mar 2022] More results on GREW are supported, and the model files are coming soon.
• [Mar 2022] Dataset GREW is supported in datasets/GREW.
• [Mar 2022] HID support is ready in datasets/HID.

Highlighted features

• Mutiple Dataset supported: OpenGait supports four popular gait datasets: CASIA-B, OUMVLP, HID, and GREW.
• Multiple Models Support: We reproduced several SOTA methods, and reached the same or even the better performance.
• DDP Support: The officially recommended Distributed Data Parallel (DDP) mode is used during both the training and testing phases.
• AMP Support: The Auto Mixed Precision (AMP) option is available.
• Nice log: We use tensorboard and logging to log everything, which looks pretty.

Model Zoo

GREW

Model	Rank@1	Configuration	Input Size	Inference Time	Model Size
Baseline	48.5	baseline.yaml	64x44	2m23s	84.12M
Baseline-ResNet9	56.0	-	64x44	-	-
GaitSet(AAAI2019)	48.4	-	-	-	-
GaitPart(CVPR2020)	47.6	-	-	-	-
GaitGL(ICCV2021)	47.3	-	-	-	-
RealGait(Arxiv now)	(54.1)	-	-	-	-

The result of GaitGL comes from RealGait now, and our reproducing results are comming soon.

OUMVLP

Model	Rank@1	Configuration	Input Size	Inference Time	Model Size
Baseline	86.7	baseline.yaml	64x44	1m13s	44.11M
GaitSet(AAAI2019)	87.2(87.1)	gaitset.yaml	64x44	1m26s	6.31M
GaitPart(CVPR2020)	88.6(88.7)	gaitpart.yaml	64x44	8m04s	3.78M
GaitGL(ICCV2021)	89.9(89.7)	gaitgl.yaml	64x44	5m23s	95.62M

CASIA-B

Model	NM	BG	CL	Configuration	Input Size	Inference Time	Model Size
Baseline	96.3	92.2	77.6	baseline.yaml	64x44	12s	3.78M
GaitSet(AAAI2019)	95.8(95.0)	90.0(87.2)	75.4(70.4)	gaitset.yaml	64x44	13s	2.59M
GaitPart(CVPR2020)	96.1(96.2)	90.7(91.5)	78.7(78.7)	gaitpart.yaml	64x44	56s	1.20M
GLN*(ECCV2020)	96.4(95.6)	93.1(92.0)	81.0(77.2)	gln_phase1.yaml, gln_phase2.yaml	128x88	47s/46s	8.54M / 14.70M
GaitGL(ICCV2021)	97.4(97.4)	94.5(94.5)	83.8(83.6)	gaitgl.yaml	64x44	38s	3.10M

The results in the parentheses are mentioned in the papers.

Note:

• All results are Rank@1, excluding identical-view cases.
• The shown result of GLN is implemented without compact block.
• Only two RTX3090 are used for infering CASIA-B, and eight are used for infering OUMVLP.

Get Started

Installation

1. clone this repo.

   git clone https://github.com/ShiqiYu/OpenGait.git
   

2. Install dependenices:

   • pytorch >= 1.6
   • torchvision
   • pyyaml
   • tensorboard
   • opencv-python
   • tqdm
   • py7zr

   Install dependenices by Anaconda:

   conda install tqdm pyyaml tensorboard opencv py7zr
   conda install pytorch==1.6.0 torchvision -c pytorch
   

   Or, Install dependenices by pip:

   pip install tqdm pyyaml tensorboard opencv-python py7zr
   pip install torch==1.6.0 torchvision==0.7.0
   

Prepare dataset

See prepare dataset.

Get trained model

• Option 1:

  python misc/download_pretrained_model.py
  

• Option 2: Go to the release page, then download the model file and uncompress it to output.

Train

Train a model by

CUDA_VISIBLE_DEVICES=0,1 python -m torch.distributed.launch --nproc_per_node=2 opengait/main.py --cfgs ./config/baseline/baseline.yaml --phase train

• python -m torch.distributed.launch DDP launch instruction.
• --nproc_per_node The number of gpus to use, and it must equal the length of CUDA_VISIBLE_DEVICES.
• --cfgs The path to config file.
• --phase Specified as train.

• --log_to_file If specified, the terminal log will be written on disk simultaneously.

You can run commands in train.sh for training different models.

Test

Evaluate the trained model by

CUDA_VISIBLE_DEVICES=0,1 python -m torch.distributed.launch --nproc_per_node=2 opengait/main.py --cfgs ./config/baseline/baseline.yaml --phase test

• --phase Specified as test.
• --iter Specify a iteration checkpoint.

Tip: Other arguments are the same as train phase.

You can run commands in test.sh for testing different models.

Customize

1. Read the detailed config to figure out the usage of needed setting items;
2. See how to create your model;
3. There are some advanced usages, refer to advanced usages, please.

Warning

• Some models may not be compatible with AMP, you can disable it by setting enable_float16 False.
• In DDP mode, zombie processes may be generated when the program terminates abnormally. You can use this command sh misc/clean_process.sh to clear them.
• We implemented the functionality about testing while training, but it slightly affected the results. None of our published models use this functionality. You can disable it by setting with_test False.
• Recommended Pytorch version: 1.6-1.8

Authors:

Open Gait Team (OGT)

• Chao Fan (樊超), [email protected]
• Chuanfu Shen (沈川福), [email protected]
• Junhao Liang (梁峻豪), [email protected]

Acknowledgement

• GLN: Saihui Hou (侯赛辉)
• GaitGL: Beibei Lin (林贝贝)
• GREW: GREW TEAM

Note: This code is only used for academic purposes, people cannot use this code for anything that might be considered commercial use.