participants
Expo Union, which was established in 2003, is actively collaborating with government agencies, associations, and private sectors to organize professional exhibitions related to ecotechnologies, such as “Circular Economy Taiwan” and “Taiwan Water.” Besides, Expo Union hosted some International Medical Conferences and governmental conferences, such as FinTech Taipei 2018(Institute for Information Industry,III), Green Energy and Frontier products Expo 2018 (Academia Sinica),World Congress on Information Technology 2017 (Institute for Information Industry,III), Asia Pacific Academy of Ophthalmology Congress 2016(The Ophthalmological Society of Taiwan). Moreover, it has organized the Marathon Expo for the masses in recent years.
A successful Expo not only requires manpower from all fields with a variety of professional abilities, but also time and effort. Site planning, Exhibition affairs, tender process are equally important. Moreover, it is essential to provide customized and a high-quality design based on customers’ needs. Expo Union has profound experience in organizing international exhibitions and has the most professional team which pays attention to every detail. Last but not least, the company aims to creates maximum values by making all-out effort, professional work division, quality communication, and careful control on cost and time to organize an excellent campaign.
Computational Intelligence Technology Center (CITC) is the first big data R&D center to bridge the big data technology across industries in Taiwan. CITC aims to develop intelligent analytics technologies to boost Artificial Intelligence and big data core capabilities for local IT software companies, also to apply intelligent analytics to help the related industries to enhance their productivity and create new business opportunities. The cross-disciplinary approach enables CITC’s core technology capabilities to provide intelligent analytics and machine learning algorithms that industries need. CITC develops an open Artificial Intelligence and big data platform to create solutions that facilitate the innovative application for service design and business model for industries.
1. 本議題重新開放報名,點選右方報名鈕即可。 2. 因運算資源有限,報名人數設有上限,額滿後關閉報名功能。 3. 提供Dockerfile,連結:https://reurl.cc/mdaj8j 。 4. 可至上述連結下載範例檔,檔名為pytorch-retinanet.zip 。 5. 上傳檔案需小於100MB。 6. 上傳即可獲得上傳禮,數量有限,送完為止。
This topic refers to the 2018 Marathon Expo, which was held in Taipei World Trade Center Exhibition Hall and lasted for three days, data contains 38,000 images taken from the camera set at the entrance.
The location and gender (male, female, and others[note]) of the visitors in the image can be distinguished rapidly by adopting the object detection technique in machine learning, so gender distribution can be known after a preliminary analysis.
For practical application, visitors’ behaviors and preferences can be analyzed by incorporating data from other devices (i.e. mobile phone). This information can be the reference of route arrangements and booth setup, exhibitors can also plan stands location and activities based on the preference of visitors.
This topic adopts a new competition mode: Participants will upload the analysis program to Aldea, the program will be executed by the platform; original data will only be kept on the platform, and the contest is going to be proceeded with no raw data released.
The purpose of this competition mode is to protect sensitive data, no data will be released on the platform, instead, the data will be read by the uploaded analysis platform. Participants can download container images and develop programs inside the container to ensure the program compatibility.
Notes: “Others” in gender category refers to cases that gender cannot be judged by their appearances, such as children and infants.
【Remote execution environment】
【Enrollment】
【Procedure】
The following awards will be given to the top three participants whose score on the Private Leaderboard meets the baseline(mAP > 0.50):
First place: 150,000 discount points of hicloud + NTD 50,000 (Tax included)
Second place: 100,000 discount points of hicloud
Third place: 50,000 discount points of hicloud
Honorable mention: 50,000 discount points of hicloud (multiple winners, depends on the final result)
Gift Redemption:
※ Participants are welcomed to test the smoothness or other related operating conditions of the platform, and are also encouraged to give feedbacks through forum discussion or admin mailbox (the forum shall have the priority). The one who gives the most valuable and the largest number of feedbacks can receive the honorable mention award!
Hicloud discount points were provided by Chunghwa Telecom
Notes:
1. Hicloud points can redeem service charge. After the discount, the price will be charged at 30% off based on the list price automatically.
2. Chunghwa Telecom deserves the right to make changes to the terms and conditions herein.
3. The example of discount points calculation can be referred to https://aidea-web.tw/computing
The activity time will be based on Taiwan Standard Time (UTC+8), and the schedule is as follows.
Time | Event |
---|---|
2019/12/25 | Competition begins |
2020/01/15 | Sample image, training set and test set announced |
2020/02/19 | Upload begins |
2020/04/29 | Competition ends |
Evaluations are conducted by adopting mean Average Precision (mAP) [1] using Intersection over Union(IoU)[2] and the threshold is 0.5. The result is True Positive (TP) if the loU of predictive object box and labeling object box is greater than 0.5, otherwise it’s False Positive (FP), and Precision index can be determined. Consequently, the system will evaluate every object to determine its AP (Average Precision) points. The mAP will then be calculated by compute the average AP points of the three gender categories. The ranking of the competitors will be based on this value. We adopt COCO API [3] to calculate the evaluated value of mAP.
Reference
[1] Average Precision (AP):
https://en.wikipedia.org/wiki/Evaluation_measures_(information_retrieval)#Average_precision
[2] intersection over union (IoU):
https://en.wikipedia.org/wiki/Jaccard_index
[3] COCO API:
https://github.com/cocodataset/cocoapi