Google’s Tensorflow is organizing a Challenge on Kaggle in partnership with Australia’s national science agency, CSIRO to help protect the great Barrier Reef. The Great Barrier Reef is a the world’s largest coral reef and home to a wide variety of flora and fauna. Recently, however, the Great Barrier Reef has been under threat from the coral-eating crown-of-thorns starfish, or COTS. Because of the overpopulation of COTS to ecologically unsustainable levels, a large-scale intervention is being planned to control the growth of COTS. This large scale intervention would however not be possible manually, therefore help is required in detection and location of COTS in images and videos by use of Artificial Intelligence and Computer Vision. To know more about the challenge, you can also check out the video below.
The Competition
In this competition, you will be provided with a training dataset that will consist of videos containing image frames that might or might not contain one or many crown-of-thorns starfish. You will have to train a model that will detect the presence and location of the crown-of-thorns starfish along with their respective confidence levels. The submissions will be tested and ranked on a hidden test dataset which will be served by an API to ensure that evaluation of the images is in the same order in which they occur in the video.
Your task is to submit the presence and locations of crown-of-thorns starfish along with their confidence levels. The solutions will be evaluated on the F2 score at different Intersection over Union(IoU) thresholds.
Timelines
The competition began on 22nd November 2021. And while the final submission deadline is on 14th February 2022, the deadline for entry into the competition and team merger is 7th February 2021. So far, there are already 134 competitors from 122 teams who have made 441 entries.
Prizes
Google is handing out big prizes for being able to accurately detect crown-of-thorns starfish in images of underwater data. The following prizes will be awarded to the top 5 finishers in the competition.
| Position | Prize |
|---|---|
| 1st | $30,000 |
| 2nd | $25,000 |
| 3rd | $20,000 |
| 4th | $15,000 |
| 5th | $10,000 |
Since the challenge is being organized for detecting the crown-of-thorns starfish in the underwater images, the models trained for these competitions will be running on an edge device for real time detection. Therefore, apart from the above mentioned prizes, Tensorflow is also handing out some additional big prizes. These prizes will be given to teams that make fastest models using Tensorflow. The top 5 teams that make the fastest models by using Tensorflow will be awarded a total of $10,000 each. To be eligible for the Tensorflow prizes, your submissions should match the following criteria-
- The submission should have a score better than or equal to the top 10% of the total teams on the leaderboard.
- Use Tensorflow 2.x on a GPU.
To be a part of the Tensorflow – Help Protect the Great Barrier Reef challenge, you just need to have a Kaggle Account. To know more about this challenge and participate in it, click here.
Good Luck!
Other Challenges
Kaggle, a subsidiary of Google is by far the biggest community of Data Science and Machine Learning practitioners. Kaggle regularly organizes competitions in various domains and on behalf of various organizations.
You can check out the blog on the Kaggle CHAII Research Challenge by Google, click here.