Li Keat

Huh, my Online R learning GitHub repo hit 500 stars! github.com/cxli233/Online…

Let's gooo!! 🌎 Try our 3D "Street View for Coral Reefs" prototype 2.0! 🪸 We did it! Today we've hit a new milestone! It's been a long journey, but we're one step closer.. We have a 50x times larger 3D Gaussian Splatting model now, covering over 1000 sq meters of corals. So stoked to tell you \o/ 🔵 Why? Coral Reefs are so precious, beautiful, incredibly complex and threatened ecosystems. They are dying fast.. But there's a way to protect and restore them! A lot of amazing people fighting for their life. We need to understand deeply how Coral Reefs function, what methods work, what don't, to coordinate precise action. They are a key to protect and restore other ecosystems. That's why we're building a digital twin of coral reef ecosystems -- our first product. 🔵 3d.wildflow.ai/C0r4Lm7 -- today you can try our second scrappy prototype (20% done) -- works best from computer. 🔵 What you see: it's a 3D "Street View" for Coral Reefs. Someone swam around a reef with a few GoPros, and you can now see high-res photorealistic 3D model of coral reef in the browser! It's a super cheap way of monitoring. Soon you will fly over square kilometres of reefs and see them in centimetre resolution. You can compare reefs in different countries and see how they evolve over time, easy way of getting rid of biases in data. 🔵 The prototype is terrible. Still work in progress (20%), just another sneak peek. We're moving fast and testing one hypothesis after another. Right now we still don't have a smooth progressive loading, there are some wacky splats here and there. 🔵 But it's already much better. This time we used 8382 images from GoPro 10 (instead of 70) and trained over 25x50m model (last time it was about 5x5 m). We respect correct geometry from Metashape (classical photogrammetry tools). Centimetre resolution! I think it's still the best in the world quality you can find. 🔵 Next steps: 3D time-series data (see how coral used to look like 3 months ago), 3D segmentation models, classifying coral species, adding other modalities (acoustics, eDNA, geospatial)... Allowing anyone to run analytics against all our data... Foundation models for biodiversity... Becoming the first generation that actually leaves behind nature better than we found it! 🪸 🔵 This version of a prototype wouldn't happen without these incredible people: - @BenWilliamsSci -- it's all Ben's fault I'm doing this now, his idea and support - @jtlrocketman -- Metashape/photogrammetry support - Greg Tkachenko -- powerful compute cluster - @rindahvida, @TimACLamont, @shebahopegrows Mars BuildingCoral team -- collecting the data, doing scientific research and restoring this reef you see - and many many more fantastic people! We've got terabytes of data from multiple orgs, now scaling the compute and processes. Soon you can play with more and more coral reefs! What a time to be alive! \o/