Dr Melissa Duncan-Schiele

One of the biggest #challenges for #research in #Maldives is that though there are numerous studies and projects on our #environment, the #data is never available. My co-authors and I have tried to understand why this is. We have summarized our findings into a #policy brief on environmental #DataSovereignty in the Maldives including #recommendations so that data collected in the Maldives isn't lost and can be harnessed to improve our society: doi.org/10.13140/RG.2.…

Our Indian Ocean marine science programme supports established researchers and has a major focus on training the next generation of ocean scientists. A big congratulations to all our recent PhD Graduates for 2024! @explorer_mel @hollystokes94 @JoannaLHarris #PhDGraduates

Introducing VGGT (CVPR'25), a feedforward Transformer that directly infers all key 3D attributes from one, a few, or hundreds of images, in seconds! No expensive optimization needed, yet delivers SOTA results for: ✅ Camera Pose Estimation ✅ Multi-view Depth Estimation ✅ Dense Point Cloud Reconstruction ✅ Point Tracking Project Page: vgg-t.github.io Code & Weights: github.com/facebookresear…

[PROTOCOL] How to 3D-map Coral Reefs with just a few GoPros 🪸🎥 A simple, low-cost way of creating permanent high-res snapshots of coral reefs for divers using 3D photogrammetry. 🔵 We were recently mapping coral reefs in East Java and I was running a mini photogrammetry workshop. A lot of you friends were reaching out about how to capture a 3D model of a coral reef so that it could be analysed and guide conservation/restoration decisions. 🔵 wildflow.ai/protocol -- here is a step-by-step tutorial explaining how to map the reef using relatively cheap equipment! 🎥 🔵 Why? Coral reefs are so precious, beautiful, incredibly complex and threatened ecosystems. We need to understand deeply how coral reefs function, what practices work, what don't, to coordinate precise actions to protect and restore them. Coral reefs are the key to protecting other ecosystems! That's why we doing it. 🔵 Sadly, In 84% of the restoration cases in Indonesia [1] people never come back to monitor their reefs. In many cases those corals would likely just die (same in other countries). The hard truth is that restoration without N years of monitoring is basically a money laundry. But we can’t blame people for that! One of the problems is the huge monitoring costs, it’s just not accessible to everyone. Once monitoring is cheap, a lot of money would be channeled towards projects that work. The whole process would be more streamlined. 🔵 There's a solution! There are thousands of scuba diving centers in the world. Almost any dive center or its visitors have an affordable camera like GoPro. It’s fairly easy to swim with a few GoPros around the reef every 6 months or a year. 🔵 The tech has moved forward very far! Why 3D Photogrammetry? It allows monitoring hectares of coral reefs in centimetre resolution: identifying corals, benthic cover (rock, sand), detecting diseases and their spread. We can track coral demography through time, calculate growth rates, carbonate budget, and more. It's not a silver bullet, but it's an awesome tool! 🔵 Fusing 3D data with other modalities like acoustics, genomics and environmental data (water temperature, salinity, dissolved oxygen etc.) unlocks a comprehensive understanding of coral ecosystem dynamics, and how we can protect and restore the reefs. 🔵 There's already a lot of organisations using this. You should give it a try as well! Watching your reef come to life in 3D is incredibly rewarding. You’ll see every coral, every nook, and every cranny. Remember, you’re not just collecting data, you’re creating a digital twin of the reef that can be studied, shared, and preserved for years to come! 🔵 Huge thanks to @JohnStratford_ , @jtlrocketman, Will Barnes, @BenWilliamsSci and other amazing people for sharing the best practices on data collection! 🔵 We could become the first generation that leaves nature behind better than we found it, if we work together! 🪸🌏