A journey to protect the Great Barrier Reef using Machine Learning
summary
A project pairs CSIRO data and an Edge ML platform with TensorFlow models trained via a Kaggle challenge to map crown of thorns starfish by GPS, aiming for real time detection. The competition draws 2,000 teams and 60,000 entries, experiments run on TPU clusters using the TensorFlow 2 Model Garden, and the deployment pipeline uses XLA, batching, quantization and anchor tuning. Data augmentation fixes overfitting and models detect many more starfish than casual human surveys.
AI-written summary · about
related talks
50mRestoring Coral Reefs at Scale: AI and the Next Generation of Conservation
47mProduction ready EdgeAI with Eclipse ioFog and Google Coral
50mML and AI at the edge with Google Coral, EdgeTPU and Cloud Platform
34mHow to Train Your Dragon and Its Friends: AI on the Edge with Eclipse Kura
29mStreamlining Competitive Data Science at CERN: Running ML...
46mCombatting illegal fishing with Machine Learning and Azure
46mIntroduction to Machine Learning
37mA Day in the Life of a Data Scientist. Conquer ML Lifecycle on Kubernetes