It's replacing a problem you can't solve (human stupidity), with one you can (a better bell).

Pretty cool though!

Bonjour, and bienvenue to the Black Mesa - pardon, la Mesa Noire - Établissement de Recherche. Please keep your hands inside le tramway at all times, and do try not to provoquer une cascade de résonance. Merci.

I just realized "Kyouryuu" is Japanese for dinosaur.

And Pfirsich is of course Peach.

Money well spent if you ask me.

Comments URL: https://news.ycombinator.com/item?id=47597935

Points: 395

# Comments: 85

But I would be surprised if car seats are the main driver here in any case.

Seems like the big AI players love buying up the good dev tooling companies.

I hope this means the Astral folks can keep doing what they are doing, because I absolutely love uv (ruff is pretty nice too).

Even the more sophisticated algorithms pretty much always do this ;-)

You are probably not interested in taking this further, but you could give the Li tree filter a try: https://pdal.org/en/stable/stages/filters.litree.html

But getting perfect segmentation is basically impossible.

It's a nice piece of work. I especially like the sections on data cleaning and registration, as that seemed to have been one of the limiting factors of the previous approaches.

I am sceptical about how accurately you can predict heights for specific trees from mono-images, but I think for cases where you just need to be right on average (e.g. biomass estimation, fuel load estimates) it's a great approach.

> We additionally release a global GeoTIFF of input image acquisition date, where pixel values encode year minus 2000 (e.g., 18.25 indicates April 2018)

That being said, I am sceptical on how accurate mono-depth models can be on a single tree basis. I would probably trust them to do large scale biomass estimates, but probably not single tree height assessments.

Maybe there are some ulterior motives, but they do also just do a little bit of "feel-good" research.

This was also in collaboration with the World Resources Institute and the University of Maryland, so it's not a 100% facebook project.

What did you do to actually count trees? Even from aerial Lidar it can be a bit finicky for closed canopies.

> CHMv2 is derived from single-date imagery, where the acquisition process selects the best available image within a target period (2017 -2020). This limits the direct use of the released CHMv2 data for attributing canopy height to a specified year of interest. To support change applications, we provide the image acquisition date associated with each prediction in the dataset metadata.

So generally a few years out of date, but the dataset is transparent about when each image was taken.

As the comments on that article rightfully point out, restricting the data analysed to 12 hours before the resolution feels like cherry picking.