REVIEW 7 cited by
Rethinking Network Design and Local Geometry in Point Cloud: A Simple Residual MLP Framework
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Point cloud analysis is challenging due to irregularity and unordered data structure. To capture the 3D geometries, prior works mainly rely on exploring sophisticated local geometric extractors using convolution, graph, or attention mechanisms. These methods, however, incur unfavorable latency during inference, and the performance saturates over the past few years. In this paper, we present a novel perspective on this task. We notice that detailed local geometrical information probably is not the key to point cloud analysis -- we introduce a pure residual MLP network, called PointMLP, which integrates no sophisticated local geometrical extractors but still performs very competitively. Equipped with a proposed lightweight geometric affine module, PointMLP delivers the new state-of-the-art on multiple datasets. On the real-world ScanObjectNN dataset, our method even surpasses the prior best method by 3.3% accuracy. We emphasize that PointMLP achieves this strong performance without any sophisticated operations, hence leading to a superior inference speed. Compared to most recent CurveNet, PointMLP trains 2x faster, tests 7x faster, and is more accurate on ModelNet40 benchmark. We hope our PointMLP may help the community towards a better understanding of point cloud analysis. The code is available at https://github.com/ma-xu/pointMLP-pytorch.
Forward citations
Cited by 7 Pith papers
-
StruMamba3D: Exploring Structural Mamba for Self-supervised Point Cloud Representation Learning
A self-supervised point cloud model that encodes spatial structure into SSM latent states and adapts state-update scale to input length achieves new SOTA on ScanObjectNN and ModelNet40.
-
Harnessing Text-to-Image Diffusion Models for Point Cloud Self-Supervised Learning
PointSD uses a frozen Stable Diffusion model, conditioned on point clouds through rendered images, to generate training targets for point cloud self-supervised learning.
-
Asymmetric Dual Self-Distillation for 3D Self-Supervised Representation Learning
AsymDSD unifies latent masked point modeling and cross-view invariance self-distillation to learn 3D representations, reporting 90.53% on ScanObjectNN and 93.72% with 930k-shape pretraining.
-
Cosmology with Topological Deep Learning
Topological neural networks using tetrahedra, clusters and hyperedges built from halo catalogs lower inference error on Omega_m by 22% and on sigma_8 by up to 60% versus graph neural networks on Quijote.
-
HiPoNet: A Multi-View Simplicial Complex Network for High Dimensional Point-Cloud and Single-Cell Data
HiPoNet combines learned feature reweighting, Vietoris-Rips complexes, and simplicial scattering transforms to classify high-dimensional point clouds, reporting top accuracy on several single-cell and spatial transcri...
-
InfoCons: Identifying Interpretable Critical Concepts in Point Clouds via Information Theory
InfoCons uses a variational information bottleneck with an attention mask to identify which points in a point cloud most influence a model's prediction.
-
Enhancing 3D Point Cloud Classification with ModelNet-R and Point-SkipNet
ModelNet-R cleans five ModelNet40 classes and Point-SkipNet is a lightweight point cloud classifier, but the reported gains over ModelNet are not proven to reflect data quality rather than a changed test set.
Discussion (0). Continue with ORCID to comment.