The code of XJTU_MM for SoccerNet2024 GameState
The method is shown in the report of Optimizing Jersey Number Recognition for Effective Player Tracking in the Game State Reconstruction
The leaderboard of SoccerNet2024 GameState
| Rank | Participant team | GS-HOTA (↑) | GS-DetA (↑) | GS-AssA (↑) | Last submission at |
|---|---|---|---|---|---|
| 1 | Constructor tech | 55.82 | 41.67 | 74.86 | 2 months ago |
| 2 | UPCxMobius | 42.19 | 30.83 | 57.78 | 2 months ago |
| 3 | XJTU_MM (JNR) | 31.17 | 19.95 | 48.74 | 2 months ago |
| 4 | VIPLab | 29.59 | 17.82 | 49.18 | 2 months ago |
| 5 | playbox x NUSG | 23.27 | 9.59 | 56.45 | 2 months ago |
| 6 | Eidos | 22.32 | 10.53 | 47.37 | 3 months ago |
| 7 | Host_17134_Team (GSR-Baseline) | 22.26 | 10.67 | 46.46 | 5 months ago |
git clone https://github.com/Xv-M-S/GameState-MM.gitCreate and activate a new environment
conda create -n tracklab pip python=3.10 pytorch==1.13.1 torchvision==0.14.1 pytorch-cuda=11.7 -c pytorch -c nvidia -y
conda activate tracklabInstall the dependencies for tracklab
cd tracklab/plugins/track
pip install -e . -i https://pypi.org/simple # note：使用pip的默认源安装
cd tracklab
pip install -e . -i https://pypi.org/simple # note 使用pip默认源安装
mim install mmcv==2.0.1Install the dependencies for sn-gamestate
cd sn-gamestate/plugins/calibration
pip install -e . -i https://pypi.org/simple # note 使用pip默认源安装
cd sn-gamestate
pip install -e . -i https://pypi.org/simple # note 使用pip默认源安装If you want to download the dataset manually, you can run the following snippet
after installing the soccernet package (pip install SoccerNet) :
from SoccerNet.Downloader import SoccerNetDownloader
mySoccerNetDownloader = SoccerNetDownloader(LocalDirectory="data/SoccerNetGS")
mySoccerNetDownloader.downloadDataTask(task="gamestate-2024",
split=["train", "valid", "test", "challenge"])
After running this code, please unzip the folders, so that the data looks like :
data/
SoccerNetGS/
train/
valid/
test/
challenge/
You can unzip them with the following command line :
cd data/SoccerNetGS
unzip gamestate-2024/train.zip -d train
unzip gamestate-2024/valid.zip -d valid
unzip gamestate-2024/test.zip -d test
unzip gamestate-2024/challenge.zip -d challenge
cd ../..External dependencies
- DATA: You will need to set up some variables before running the code in soccernet.yaml(sn_gamestate/configs/soccernet.yaml)
data_dir: the directory where you will store the different datasets (must be an absolute path !). If you opted for the automatic download option,data_dirshould already point to the correct location.
- MODEL: Download the pretrained model weights here and put the "pretrained_models" directory under the main project directory (i.e. "/path/to/tracklab/pretrained_models/reid").
- YoloModel: Dowlaod the pretrained YOLOv8 model weights here and put the "yolov8x6.pt" file under the main project directory (i.e. "/path/to/tracklab/pretrained_models/yolo").
cd sn-gamestate
python -m tracklab.main -cn soccernet