Data

Tools

Indexes

Applications

Experiments

Open Source Science

automated-hit-frame-detection-for-badminton-match-analysis

https://github.com/arthur900530/automated-hit-frame-detection-for-badminton-match-analysis

Science Score: 54.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ✓
    CITATION.cff file
    Found CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ○
    DOI references
  • ✓
    Academic publication links
    Links to: arxiv.org
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (6.7%) to scientific vocabulary
Last synced: 10 months ago · JSON representation ·

Repository

Basic Info
  • Host: GitHub
  • Owner: arthur900530
  • License: mit
  • Language: Python
  • Default Branch: master
  • Size: 1.21 GB
Statistics
  • Stars: 13
  • Watchers: 1
  • Forks: 2
  • Open Issues: 1
  • Releases: 0
Created about 3 years ago · Last pushed almost 3 years ago
Metadata Files
Readme License Citation

README.md

Automated Hit-frame Detection for Badminton Match Analysis

This repo contains official implementation of Automated Hit-frame Detection for Badminton Match Analysis. The main contributions of our work are to: - Utilize the interchange of shot angle to trim rally-wise frames from raw badminton video. - Proposed a novel transformer that predicts shuttlecock direction sequences based on player keypoint sequence. - Designed and developed the first-ever automated hit-frame detection tool to bridge the gap between raw badminton videos and analyzable data.

Paper: https://arxiv.org/abs/2307.16000

Checkpoints: https://drive.google.com/drive/folders/1v-uejba2ljNRUPaSAR-9u-a9GY1mD4pu?usp=sharing

Datasets: https://drive.google.com/drive/folders/1oZf7U5qx70YuNZX5Df6y1vJ_WwrMEprD?usp=sharing

Environment Requirements

opencv-python == 4.7.0.72 python == 3.8 scikit-learn == 1.0.2 torch == 2.0.1 torchvision == 0.15.2 yaml

YAML Parameters

- model: - sacnn_path: Path to SA-CNN's weight Default is './models/weights/sacnn.pt'. - court_kpRCNN_path: Path to Court Keypoint-RCNN's weight Default is './models/weights/court_kpRCNN.pth'. - kpRCNN_path: Path to Keypoint-RCNN's weight Default is './models/weights/kpRCNN.pth'. - opt_path: Path to transformer model's weight Default is './models/weights/OPT_16_head_dp.pt'. - scaler_path: Path to data scaler's file Default is './models/weights/scaler.pickle'. - sa_queue length: Length of the sa_queue Default is 5. - video_directory: Directory with unresolved videos Default is '../videos'. - video_save_path: Directory to store resolved videos Default is '../outputs/videos'. - joint_save_path: Directory to store player joints and frame info Default is '../outputs/joints'. - rally_save_path: Directory to store rally-wise info Default is '../outputs/rallies'.

Run the Code

python main.py

Options

--yaml_path STR Path to AI Coach setting yaml. Default is "../configs/ai_coach.yaml".

Owner

  • Name: Arthur Chien
  • Login: arthur900530
  • Kind: user

Strive for excellence and be patient

Citation (CITATION.cff) 

cff-version: 1.2.0
message: "If you use this software, please cite it as below."
authors:
- family-names: "Chien"
  given-names: "Yu-Hang"
  orcid: "https://orcid.org/0000-0002-7334-327X"
- family-names: "Yu"
  given-names: "Fang"
  orcid: "https://orcid.org/0000-0002-2776-9624"
title: "Automated-Hit-frame-Detection-for-Badminton-Match-Analysis"
version: 1.0.0
doi: 10.5281/zenodo.1234
date-released: 2023-07-23
url: "https://github.com/arthur900530/Automated-Hit-frame-Detection-for-Badminton-Match-Analysis"

GitHub Events

Total
  • Watch event: 9
  • Fork event: 2
Last Year
  • Watch event: 9
  • Fork event: 2