IIT GOA

INDIAN INSTITUTE OF TECHNOLOGY GOA
भारतीय प्रौद्योगिकी संस्थान गोवा

Express Analytics-IIT GOA Competition in Artificial Intelligence (AI) - 2019

The problem that Express Analytics (EA) and IIT GOA has set forward for our jointly organized Competition in Artificial Intelligence (AI)-2019 to resolve is as follows:

The Kinetics Human Action Video Data Set [1,2] is a data set of about 500,000 Youtube Video clips that depict persons engaging in physical actions; there are two types of videos, one that encompasses videos depicting one of 400 unique actions (Kinetics-400), and the other that features videos depicting one of 600 actions (Kinetics-600). These actions can vary from simple motions, like applauding, to something that is complex, such as wrapping a present or weaving a basket. For this project, you will be asked to work with dataset Kinetics 400.

This project expects you to develop a machine learning algorithm that can, after being trained on a training set derived from Kinetics 400 [1], determine what action is being depicted in a given test video. Training, validation, and test sets are given in [1].

The challenges you will likely encounter are as follows, and will better explain why this development is both so important and so valuable:

  1. Unlike image recognition, action recognition on a video is significantly more complex and is based on accurate recognition of movements as opposed to being able to match color or shape cues.
  2. Video recognition has a high quotient of use cases, such as to identify and flag offensive videos, tag unique persons for customer tracking in retail applications, and recommend videos based on content.
Competition Rules
  1. Team sizes are restricted to 1-3 members.
  2. All members in a team should be students currently enrolled in any degree program, engineering or otherwise, in any recognized educational institution in India.
  3. Teams should register before submitting their solutions. Registration link: https://forms.gle/tmfF8CmQqFwe6yze7
  4. Registration is allowed up to two days before the submission date for the solution.
  5. All teams have to abide by the rules, conditions and decisions taken by the organizers of the competition.
Submission Guidelines

You should follow the submission guidelines of Kaggle as given at the link below to submit your complete solution:

https://www.kaggle.com/WinningModelDocumentationGuidelines

Schedule of Major Events:

Below is a list of the important dates of the competition, which are hard deadlines. Please make sure you submit your team’s final algorithm and associated files as an organized .zip archive by the deadline date. We do not consider late submissions.

Final submission date: August 31, 2019 11:59 PM

Winning team will be announced on: October 1, 2019

Evaluation Criteria

All entries of the competition that are received by (insert datetime) will be judged on a rubric that will not be disclosed to competitors. After evaluation by an expert committee, we will select the winner for each of the following categories:

  1. Most Well-Designed Solution -- This award will go to the team that has designed the algorithm that is most user-friendly and is best designed for customer-facing applications. Teams that seek to win this prize should consider building a high-quality graphical user interface for their algorithm in addition to the accuracy of the solution.
  2. Fastest Algorithm Processing Time and Most Accurate Algorithm -- This award will go to the team that, when run, completes training and testing in the smallest amount of time; In addition to the overall time the algorithm takes, the amount of time it takes the algorithm to be trained on the Kinetics-400 training set and the amount of time the algorithm takes to produce predictions for the Kinetics-400 test set will be considered.
  3. Best Overall Performance Combining 1 and 2 Above
Prizes/Awards:
Contact Us

If you have any questions related to the competition, you can contact

References
Disclaimer

Express Analytics and IIT Goa reserve the rights to modify and alter any information related to the competition at any point in time, including the number of awards/prizes to be given for different evaluation criteria. No communication in this regard will be entertained.