yubin
Re:Denser 3D Point Clouds in OpenSfM
on Monday October 31, 2016 @10:24AM (#12836929)
Both VisualSfM and OpenSfM use similar methodology (incremental structure from motion) to reconstruct 3D scene points and camera positions from photos.
And yes, you can build your own tools (even commercially) with OpenSfM.
Re:Denser 3D Point Clouds in OpenSfM
on Monday October 31, 2016 @10:18AM (#12836876)
Exactly, the convnet is applied on 2d pixels instead of on 3D point clouds. If one have labelled 3D points (e.g. if a point is part of a car or not), one can train a network for recognition directly on 3D points.
Re:Denser 3D Point Clouds in OpenSfM
on Monday October 31, 2016 @10:15AM (#12836844)
Yes. We can apply different heuristics to remove outlier points. One interesting way to do it will be to train a neural network to refine the dense point cloud.