Hacker News Reader

Hacker News

Comment by dang | Hacker News Reader

Full thread

dang

https://arxiv.org/abs/2212.08751 (via https://news.ycombinator.com/item?id=34060986)

https://techcrunch.com/2022/12/20/openai-releases-point-e-an... (via https://news.ycombinator.com/item?id=34069231)

https://twitter.com/drjimfan/status/1605175485897625602 (via https://news.ycombinator.com/item?id=34068271)

(but no meaningful comments at those other threads)

View on HN

Maybe the lack of commenter enthusiasm is because point clouds are fairly specialized. Most people don’t have interesting point cloud data lying around to test this with, or the means to capture such data.

3D sensors are slowly but surely becoming more common. The iPhone Pro series has one, and AR hardware designs tend to include these capabilities. So this model synthesis seems a bit ahead of the curve, in a good way.

Agree with you that point clouds aren't mainstream at all and most people aren't sure what they'd use them for.

I think the premise of this is text-to-3D, and that because it's quicker generations you don't really need anything besides a GPU to start playing around with it.

anyone with a recent (last five years) iphone or ipad has the means to generate point cloud data using the depth sensors.

Do you have an app to recommend? And does it work well on small objects? The apps I tried were not very impressive.

Sorry I don’t know any store apps for such, my only experience is through personal corespondance/demos with/by developers experimenting with the hardware feature and sdk. Quick googling turns up some contenders but I can’t vouch for them:

https://apps.apple.com/ca/app/point-cloud-ar/id1435700044

https://apps.apple.com/ca/app/point-precise/id1629822901

Polycam https://apps.apple.com/us/app/polycam-lidar-3d-scanner/id153... is good for outdoor scanning in my limited experience.

It generates nerfs (which have some advantages as well as disadvantages depending on application) but Luma AI is arguably SotA for photogrammetry on iPhones.

Theoretically this is the case, but in my testing with several apps on an iPhone 13 Pro the results are underwhelming.

> Maybe the lack of commenter enthusiasm is because point clouds are fairly specialized.

Please correct me if I am wrong.

The pointcloudtomesh notebook seems to be be able to output something could be converted for 3d printing purposes.

I haven’t yet attempted to do so, but that does seem like an exciting and general purpose use case.

There is a lot of publicly available realtime satellite data that nicely converts to point cloud based on cloud top temperature.

Can't you easily convert point clouds to polygons? If you know the sampling frequency, it's trivial to identify holes/edges/faces and just convert it to a rough polygonal model. Then you can run edge collapses with a desired error to make it smoother.

Reply on news.ycombinator.com