PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
read the original abstract
Point cloud is an important type of geometric data structure. Due to its irregular format, most researchers transform such data to regular 3D voxel grids or collections of images. This, however, renders data unnecessarily voluminous and causes issues. In this paper, we design a novel type of neural network that directly consumes point clouds and well respects the permutation invariance of points in the input. Our network, named PointNet, provides a unified architecture for applications ranging from object classification, part segmentation, to scene semantic parsing. Though simple, PointNet is highly efficient and effective. Empirically, it shows strong performance on par or even better than state of the art. Theoretically, we provide analysis towards understanding of what the network has learnt and why the network is robust with respect to input perturbation and corruption.
This paper has not been read by Pith yet.
Forward citations
Cited by 8 Pith papers
-
Patch Hierarchical Attention Transformer for Efficient Particle Jet Tagging
PHAT-JeT combines geometric message-passing with hierarchical patch attention to reach state-of-the-art accuracy and background rejection among resource-constrained jet tagging models on four benchmarks.
-
StereoPolicy: Improving Robotic Manipulation Policies via Stereo Perception
StereoPolicy fuses stereo image pairs via a Stereo Transformer on pretrained 2D encoders to boost robotic manipulation policies, showing gains over monocular, RGB-D, point cloud, and multi-view methods in simulations ...
-
SegviGen: Repurposing 3D Generative Model for Part Segmentation
SegviGen shows pretrained 3D generative models can be repurposed for part segmentation via voxel colorization, beating prior methods by 40% interactively and 15% on full segmentation using only 0.32% of labeled data.
-
Dual Grid Net: hand mesh vertex regression from single depth maps
Dual Grid Net is a two-stage FCN that regresses 3D hand mesh vertices and dense correspondences from single depth maps, achieving SOTA keypoint accuracy on NYU under supervision and competitive results via self-superv...
-
OPTNet: Ordering Point Transformer Network for Post-disaster 3D Semantic Segmentation
OPTNet adds a learnable ordering module with self-supervised loss to Point Transformers for improved efficiency and accuracy in post-disaster 3D semantic segmentation on the 3DAeroRelief dataset.
-
From Spherical to Gaussian: A Comparative Analysis of Point Cloud Cropping Strategies in Large-Scale 3D Environments
Gaussian and related cropping strategies for point cloud subclouds improve 3D neural network performance over spherical cropping on large outdoor scenes.
-
From Spherical to Gaussian: A Comparative Analysis of Point Cloud Cropping Strategies in Large-Scale 3D Environments
Gaussian and linear cropping strategies for large point clouds improve 3D neural network performance over spherical crops, especially in outdoor scenes, and achieve new state-of-the-art results.
-
Enwar 3.0: An Agentic Multi-Modal LLM Orchestrator for Situation-Aware Beamforming, Blockage Prediction, and Handover Management
Enwar 3.0 is an LLM-orchestrated framework that uses a sensor degradation classifier and context-aware agent coordination to achieve over 88% beam selection accuracy, 98% blockage F1-score, and 87% reasoning correctne...
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.