This repo extends the PointNet++ model for semantic segmentation of buildings. The model is trained on a portion of the 2022 Vancouver LiDAR Dataset whereby the points are filtered by their proximity ...

Abstract: Sample-efficient neural architecture search (NAS) techniques have advanced rapidly. Two lines of methods, namely neural predictor and sequential search, have shown promising performance in ...

In this project, we compare MCMC methods with diffusion based methods, in particular we use pre-conditioned Crank-Nicholson (pCN) with Metropolis Hastings with TV Prior. In this work, we extend ...