undefined | Better HN

0 pointstimhigins9mo ago0 comments

This has been researched quite a lot with the adjacent task of structure from motion (2d videos) - see NeRFs, Gaussian Splatting - and now there are a few models that do 3d from single 2d images fairly well.

0 comments

mhl479mo ago

And interestingly it's still not even close to what a good human artist could do in this task having the same data.

j / k navigate · click thread line to collapse

0 pointstimhigins9mo ago0 comments

This has been researched quite a lot with the adjacent task of structure from motion (2d videos) - see NeRFs, Gaussian Splatting - and now there are a few models that do 3d from single 2d images fairly well.

0 comments

mhl479mo ago

And interestingly it's still not even close to what a good human artist could do in this task having the same data.

j / k navigate · click thread line to collapse