This is a Kaggle Competition 🐳Happywhale

Descriptions

Algorithms developed in this competition will be implemented in Happywhale, a research collaboration and citizen science web platform. Its mission is to increase global understanding and caring for marine environments through high quality conservation science and education. Happywhale aims to make it easy and rewarding for the public to participate in science by building innovative tools to engage anyone interested in marine mammals. The platform also serves the research community with powerful collaborative tools.

The photos of whales are margined with extensive noisy water waves. As you can see, the middle one is the photo after segmentation.

My works

  • Rank: 303/1588. Win silver medal notebook from my open-source PyTorch baseline of this competition.
  • Identify the whale individual only by pictures of their tails, fins, or heads. Apply U-Net to extract the fins out of the noisy background and thus make a purer dataset for this competition.

Contribution

🐳✨Background Remove Tutorial✨

During the competition, I publish a dataset, which contains the images after segmentation.