Registration
00

day

00

hr

00

min

00

sec

Time to team-up deadline

Topic provider

Competition Guidance Unit: Department of Information and Technology Education, Ministry of Education

Competition planning unit: Office of AI and Annotation Data Collection Project, Ministry of Education

Topic provider: Network Optimization Laboratory, Department of Computer Science and Information Engineering, National Yang Ming Chiao Tung University

Competition co -organizers : National Yang Ming Chiao Tung University Precision Sport Science Team, National Central University Department of Computer Science and Information Engineering, National Taiwan University Department of Information Engineering, Chailease Finance Co., Information Industry Development Association, Industrial Technology Research Institute

2023/03/02 關於報名欄位顯示「額滿」說明

由於此競賽須由 AI CUP 官網進行報名，為避免參賽者困惑，故 AIdea 競賽頁面之報名欄位將顯示「額滿」。 再次提醒，欲參賽者請透過：https://go.aicup.tw/ 進行報名，謝謝。

Introduction

According to global statistics, there are approximately 2.2 billion badminton players worldwide and more than 3 million in Taiwan. This single sport is ranked second in terms of national popularity. In recent years, badminton players have achieved outstanding performances in international competitions, gradually increasing the public attention.

In terms of badminton skills and tactics analysis, our team has proposed a match shuttlecock recording format and developed a computer vision assisted quick shuttlecock labeling program to initiate the research of badminton big data. Although many computer-assisted techniques have been used, manual shuttlecock labeling still requires manpower and time, especially for technical data identification, which requires badminton experts to perform. Through this competition, we hope to invite machine learning, image processing, and sports science expertise to develop an automatic shuttlecock labeling model with a high recognition rate, making the massive badminton information collection possible, and thus popularizing the research and application of badminton tactics analysis.

For any related inquiries, please contact: evawang.cs11@nycu.edu.tw
Competition forum: 2023 AI CUP：教電腦、看羽球、AI CUP 實戰人工智慧

This topic is not open for direct registration on the AIdea platform. Participants who wish to participate should register through the AI CUP registration system. If it is the first time for a participant to use the AI CUP registration system, please refer to the link for the AI CUP registration system process.

After completing the registration, please fill out the pre-test questionnaire.

 

Participants

  1. Registration is open to all students (e.g. high school students, college students, graduate students, etc.) and members of the general public.
  2. The competition is divided into student and public groups. The student group requires all team members to be students; if any member of the team is not a student, the team will be classified as the public group. Only student group teams will be included in the overall competition ranking.

Prize

Award Prize Money
Champion1 team70,000 TWD
Runner-up1 team50,000 TWD
Third place1 team30,000 TWD
Best Paper Award1 team6,676 TWD
Honorable Mention14 teams6,666 TWD
  1. All members of the student team must be current students at the time of registration, and relevant certificates must be attached when the awards are presented.
  2. The prize will be distributed by remittance in NT dollars, and the winning team should pay the relevant income tax in accordance with the tax laws of the Republic of China.
  3. The winning team must submit their original code and final report that can reproduce the model before the specified deadline to verify the results.
  4. The instructor/mentor cannot register as a contestant with the student team; they can only assist the contestants as an instructor. If the competition team has an instructor/mentor, please fill in the relevant fields in the registration system and fill in the information of the instructor/mentor (including name, school/company name, department/title, e-mail address) in the final report. The host will contact the instructor/mentor to fill in the proof of guidance according to the provided e-mail address information. The instructor/mentor registration is optional; if there is none, it can be left blank.
  5. The top 13 student teams will receive a certificate of honor from the Ministry of Education of ROC. The personal information of the winning team members will be based on the information filled in the final report and will be used as the information for the production and distribution of the awards. The Best Paper Award will be based on scoring the uploaded report and program code. The Best Paper Award can be awarded concurrently with other awards.
  6. Teams whose rankings are in the top 25% and exceed the baseline (including the public group) will be awarded an electronic certificate of honor from the Ministry of Education Artificial Intelligence Competition Project Office after review by the judges. This award will not be awarded together with the previous award.
  7. Any member of the winning team of the student teams who have reached the top three ranking three times or above in a previous "Ministry of Education's National College Artificial Intelligence Competition ( AI CUP )" organized by the "Ministry of Education's Artificial Intelligence Competition and Labelling Data Collection Project Office", will not be restricted from continuing to participate in the competition. The award is cumulative from the fall competition of 2022. If future competitions meet the above conditions, the team will only receive a certificate of honor from the Ministry of Education and no prize money. The prize money will be transferred to the second-place team, and the two teams will be tied for the same place.
  8. The number of awards will be adjusted according to the number of entries, scores, and quality of report writing. Entries that do not meet the standards will be selected for deficiency or lack of entry by the selection judges.
  9. Winning teams must submit a complete implementation program and design documents to the host within the designated deadline after the competition is over. Reports containing errors or omissions after the deadline, or those submitted after the deadline, will not be reviewed. If necessary, the host may require the contestants to help set up the environment to reproduce. The contestants cannot refuse, otherwise they will be considered as not passing the review.
  10. The winning teams must designate at least one representative to attend the subsequent award activities according to the notification of the host. If there are no teammates to help, family and friends can be invited to attend the award ceremony and make presentations.
  11. The top three award-winning teams must agree to make their code open source, and those who do not agree to making their code open source can only be classified as excellent entries at the review stage at best.

Activity time

DateEvent Schedule
2023/03/01 ( Wed )Registration opens
2023/03/15 ( Wed )Training and test data sets open for download
2023/03/22 ( Wed )Open for result uploading and
public leaderboard scores announcement
2023/05/02 ( Tue )
11:59:59 am		Registration closes
2023/05/09 ( Tue )
11:59:59 am		Private test data set open for download and
public + private answer uploading,
but only public leaderboard scores will be announced
2023/05/16 ( Tue )
23:59:59 pm		Result uploading closes
2023/05/17 ( Wed )
17:00:00 pm		Private leaderboard scores announcement and
open for uploading reports and codes
2023/05/24 ( Wed )
23:59:59 pm		Report and program code uploading deadline
2023/06/09 ( Fri )Final results announcement
2023/07 ( Tentative )Award ceremony and prize money distribution,
all arranged by the Project Office

Evaluation Criteria

  1. Each single video clip is scored as a separate question, with a total score of 1 point per question. First, compare the number of shots in the score video, with each row of the CSV file considered as one shot: if the predicted number of shots is different from the true value, the question will be scored as zero; if the number of shots is the same, 0.1 points will be awarded and proceed to the next stage to compare the content of the columns, with the highest score for column content being 0.9 points.
  2. If the number of shots is correct, compare each shot in sequence, with each shot scored according to the total score of the columns, and the average of each shot's score as the score for column content. First check the HitFrame column: if the prediction error is greater than 2 frames compared to the true value, the shot will get 0 points and proceed to the next shot for scoring; if the prediction error is not greater than 2 frames, 0.1 points will be given and proceed to compare other columns in the same shot.
  • Hitter : 0.1 points if correct, 0 points otherwise.
  • BallHeight : 0.1 points if correct, 0 points otherwise.
  • Landing: If the prediction error is not greater than 6 pixels in the Euclidean distance, it is considered correct and is awarded 0.1 points; otherwise, 0 points.

$$ \sqrt{(x_{gt}-x_{pred})^2+(y_{gt}-y_{pred})^2}<6 $$

  • HitterLocation : If the prediction error is less than 10 pixels ( Euclidean distance ) , it is considered correct with 0.05 points, otherwise 0 points.

$$ \sqrt{(x_{gt}-x_{pred})^2+(y_{gt}-y_{pred})^2}<10 $$

  • DefenderLocation : If the prediction error is less than 10 pixels ( Euclidean distance ) , it is considered correct with 0.05 points, otherwise 0 points.

$$ \sqrt{(x_{gt}-x_{pred})^2+(y_{gt}-y_{pred})^2}<10 $$

  • Backhand : 0.05 points if correct, 0 points otherwise .
  • RoundHead : 0.05 points if correct, 0 points otherwise .
  • BallType : 0.2 points if correct, 0 points otherwise.
  • Winner : 0.1 points if correct, 0 points otherwise . (Note : If it is not the last shot but is filled in, it will be considered incorrect.)

Assuming that there are $R$ rally videos in the data set, and the $i$th video has $S_i$ shots, the scoring formula is given below:

$$ \frac1R\sum_{i=1}^R1_{S_i=S_i^{pred}}(0.1+ASS_i) $$

in which $ASS_i$ (the Average Shot Score of content of the $i$-th rally video) is given by:

$$ ASS_i=\frac1{S_i}\sum_{j=1}^{S_i}1_{\vert HitFrame_j-HitFrame_j^{pred}\vert\leq2}SS_j $$

with $SS_j$ (the $j$-th Shot Score) given by:

$$ \begin{array}{lcl}SS_j&=&0.1+0.1\times1_{Hitter_j=Hitter_j^{pred}}\\&&+0.1\times1_{BallHeight_j=BallHeight_j^{pred}}\\&&+0.1\times1_{\left\|Landing_j-Landing_j^{pred}\right\|<6}\\&&+0.05\times1_{\left\|HitterLocation_j-HitterLocation_j^{pred}\right\|<10}\\&&+0.05\times1_{\left\|DefenderLocation_j-DefenderLocation_j^{pred}\right\|<10}\\&&+0.05\times1_{Backhand_j=Backhand_j^{pred}}\\&&+0.05\times1_{RoundHead_j=RoundHead_j^{pred}}\\&&+0.2\times1_{BallType_j=BallType_j^{pred}}\\&&+0.1\times1_{Winner_j=Winner_j^{pred}}\end{array} $$


Rules

  1. Test dataset predictions are uploaded up to five times a day, with the scores obtained from the last upload ranking. When submitting the private answer, the answer (public + private) must be submitted together with the public test dataset prediction result.
  2. Teams are not allowed to register multiple accounts to participate in the same competition, otherwise they will be disqualified. Individual members of the team submitting results with their own personal accounts will also count as participating with multiple accounts.
  3. Team Competition Mode:
    1. Teams enroll according to the competition group, with a maximum of 5 members.
    2. Each team selects a captain, and all rights and obligations (including notifications sent by the organizer and the distribution of awards, etc.) shall be notified and accepted by the captain.
    3. After forming a team, new members cannot be added.
    4. Once you join a team, you cannot change teams .
  4. Any member of the winning team of the student teams who have reached the top three ranking three times or above in a previous "Ministry of Education's National College Artificial Intelligence Competition ( AI CUP ) " organized by the "Ministry of Education's Artificial Intelligence Competition and Labeling Data Collection Project Office", will not be restricted from continuing to participate in the competition. The award is cumulative from the fall competition of 2022. If future competitions meet the above conditions, the team will only receive a certificate of honor from the Ministry of Education and no prize money. The prize money will be transferred to the second-place team, and the two teams will be tied for the same place.
  5. Teams can use additional resources such as additional images and packages to improve the model training results, but they cannot manually annotate the test dataset (Test Data) in an artificial manner. If additional resources are used, a source explanation should be provided in the written report submitted.
  6. The data, technology and code used in the participating works belong to the participating teams. Any third party claiming infringement of intellectual property rights or other violations shall be dealt with by the participating teams themselves. The violation of the intellectual property rights of others, if verified by the organizer, will disqualify the team from the competition and revoke their award eligibility, and the participating teams will bear the relevant legal responsibilities themselves.
  7. To ensure the fairness of the competition, the organizers will require the award-winning teams to submit a final report, including but not limited to (1) members, (2) model algorithm explanation, (3) original code and explanation, (4) additional resources. The submitted code should be successfully executed in Google Colaboratory to verify the results. In case of fraud, plagiarism or cheating, the organizer may cancel the contestant's eligibility and reclaim the prize. The contestant will bear the legal responsibility.
  8. The intellectual property rights of the competition results belong to the contestant's team. The related matters of copyright authorization, patent application, technology transfer, and rights distribution should be handled in accordance with the relevant laws and regulations
  9. The award-winning team shall pay relevant income tax in accordance with the tax laws of the Republic of China.
  10. The organizer reserves the right to adjust the dataset during the competition if necessary.
  11. In the event of any of the following circumstances, the organizer may disqualify the contestant or revoke their award eligibility without informing the contestant:
    1. There has been specific evidence that the team they belong to has committed plagiarism, cheating, or fraud.
    2. There has been specific evidence that the team they belong to has infringed on the intellectual property rights of others.
    3. There has been specific evidence that the team they belong to has attacked the Leaderboard system.
    4. There has been specific evidence that the team they belong to has affected other participating teams in ways that led to unfair cases.
  12. Teams that violate the rules of this competition will be disqualified. If awards have been given, the awards will be revoked, and the prize money and certificates will be retrieved.
  13. The recipients of the Ministry of Education’s prize money must be students with Taiwan citizenship who are enrolled in school at the time of registration. The prize must also be awarded to the entire team, and relevant proof must be presented when the prize is given. If there are teams composed of members of the public in this competition, they can receive an electronic certificate from the Office of the Artificial Intelligence Competition Project of the Ministry of Education.
  14. The instructor/mentor cannot register as a contestant with the student team; they can only assist the contestants as an instructor. If the competition team has an instructor/mentor, please fill in the relevant fields in the registration system and fill in the information of the instructor/mentor (including name, school/company name, department/title, e-mail address) in the final report. The host will contact the instructor/mentor to fill in the proof of guidance according to the provided e-mail address information. The instructor/mentor registration is optional; if there is none, it can be left blank.
  15. The award-winning instructor/mentor can only receive a certificate from the Ministry of Education, but not a bonus from the Ministry of Education, and the instructor/mentor filled in the student's final report shall be the one acknowledged.
  16. All other matters not specified herein shall be mutually discussed and agreed upon by the parties in accordance with the existing circumstances.
  17. All entrants shall be deemed to have read and fully agreed to all the regulations of this event.
  18. The organizer reserves the right to the final interpretation of this event in case of any dispute.

Topics Topic List Seminars AI CUP 2022 CTSP-Smart Manufacturing ASVDA Career Playground
About About Us Our Team Computing Resources Privacy policy FAQ System Status
Contact

For further suggestions or questions
admin.AIdea@itri.org.tw

繁體中文 English

Copyright © 2017–2023 Industrial Technology Research Institute（AIdea platform v1.5.0）