In November 2021, Zindi and DeepMind, and several marine conservation organizations came together to design a challenge that would aid conservation efforts by using machine learning to distinguish between turtles of the same species. The aim of the competition was to build a machine learning model that would identify individual sea turtle faces; this technology will make future identification of turtles easier, contributing studies of population dynamics and other turtle and marine conservation efforts.
“Overall, we are thrilled by the level of engagement from the community and impressed by their technical innovation on this challenging task,” says Annette of the DeepMind Turtle Recall Team.
DeepMind is a team of scientists, engineers, and machine learning experts working together to advance the state of the art in artificial intelligence (AI), committed to solving intelligence, to advance science and benefit humanity.
“Our turtle conservation partners say that this level of prediction will be immediately valuable to them in the field, meaning that the results will have a real and immediate impact on wildlife conservation,” says Annette.
The challenge participants had a positive experience in helping build world-first models that can recognize turtle faces. Zindian and 2nd placed winner Flame Turbo says: “We really enjoyed the competition, and particularly spent a lot of time researching how state-of-the-art facial recognition systems are built. We learnt a lot and continued trying ideas until the last minute. We really appreciate Zindi and DeepMind for organizing this amazing challenge.”
“We were thrilled to have received thousands of submissions from across the world, and in particular from 28 African countries. Overall, we found this experience to be very engaging and beneficial and more importantly, valuable for the broader machine learning community. Receiving an ML solution with a 98% score was absolutely incredible, and this will be beneficial to us and other organizations that want to use facial recognition systems for conservation,” says Annette.
The solution provided by the three top winners will help marine conservation organizations establish systems that will not only help identify turtles, but aid in the conservation of all the marine life.
The facial recognition machine will be deployed by 2023. Zindi and DeepMind eagerly await the deployment of the system and the impact that it will drive in African marine conservation.
We thank Local Ocean Conservation (LOC) for providing us with the data to create a machine learning tool that will help in the conservation of marine life, and DeepMind for their support of this important initiative.