ConvPoint: Continuous convolutions for point cloud processing

TL;DR: PAConv as mentioned in this paper constructs the convolution kernel by dynamically assembling basic weight matrices stored in Weight Bank, where the coefficients of these weights are self-adaptively learned from point positions through ScoreNet.

TL;DR: This work aims at facilitating research on 3D representation learning by selecting a suite of diverse datasets and tasks to measure the effect of unsupervised pre-training on a large source set of 3D scenes and achieving improvement over recent best results in segmentation and detection across 6 different benchmarks.

TL;DR: In this article, a triplet of architecture, source dataset, and contrastive loss for pre-training is used for 3D point cloud point cloud segmentation and detection in indoor and outdoor, real and synthetic datasets.

TL;DR: Li et al. as discussed by the authors proposed a self-training approach, in which they iteratively conduct the training and label propagation, facilitated by a graph propagation module and adopt a relation network to generate the per-category prototype and explicitly model the similarity among graph nodes to generate pseudo labels.

TL;DR: In this article, the authors explore ways to scale up networks in ways that aim at utilizing the added computation as efficiently as possible by suitably factorized convolutions and aggressive regularization.

TL;DR: An automatic differentiation module of PyTorch is described — a library designed to enable rapid research on machine learning models that focuses on differentiation of purely imperative programs, with a focus on extensibility and low overhead.

TL;DR: A new neural network model, called graph neural network (GNN) model, that extends existing neural network methods for processing the data represented in graph domains, and implements a function tau(G,n) isin IRm that maps a graph G and one of its nodes n into an m-dimensional Euclidean space.

TL;DR: This work proposes a new neural network module suitable for CNN-based high-level tasks on point clouds, including classification and segmentation called EdgeConv, which acts on graphs dynamically computed in each layer of the network.