Global Forest Canopy, 3D Meshes, and Image Optimization

1. Seeing the Forest AND the Trees

I saw a very cool new geospatial data product from Meta and the World Resources Institute. It’s a high-resolution global canopy height layer at 1-m resolution (publication here).

“The model was trained on 18 million satellite images (0.5m natural color imagery from Maxar Technologies), encompassing more than a trillion pixels, from across the globe allowing the detection of single trees at a global scale. In an effort to advance open source forest monitoring, all canopy height data and artificial intelligence models are free and publicly available”

Above is a demo from the Google Earth Engine site to view the layer, showing the satellite imagery and canopy height layer above. I was even able to pick out individual trees around my house!

2. Creating a 3D Mesh from a Single Photo

InstantMesh is a feed-forward framework for efficient 3D mesh generation from a single image based on the LRM/Instant3D architecture.

Here I generated an image of a hummingbird from DALL-E and fed it through InstantMesh. The result is pretty impressive!

This is pretty mind-blowing to be able to do this with a single image. I could see this making 3D stuff a lot more approachable in the future.

3. Open Source Image Optimization with imgproxy

imgproxy is an open source image optimization platform, and can be deployed with Docker. This looks like a really cool alternative to something like Cloudinary. There is also a pro version with advanced features, of course.