3D Point Cloud Linear Classification On

Metrics

Overall Accuracy

Results

Performance results of various models on this benchmark

Model Name	Overall Accuracy	Paper Title	Repository
IAE (DGCNN)	92.1	Implicit Autoencoder for Point-Cloud Self-Supervised Representation Learning
I2P-MAE	93.4	Learning 3D Representations from 2D Pre-trained Models via Image-to-Point Masked Autoencoders
AdaCrossNet	91.4	AdaCrossNet: Adaptive Dynamic Loss Weighting for Cross-Modal Contrastive Point Cloud Learning
ReCon	93.4	Contrast with Reconstruct: Contrastive 3D Representation Learning Guided by Generative Pretraining
OcCo	89.2	Unsupervised Point Cloud Pre-Training via Occlusion Completion
CrossPoint	91.2	CrossPoint: Self-Supervised Cross-Modal Contrastive Learning for 3D Point Cloud Understanding
MAE-VAE	88.4	Multi-Angle Point Cloud-VAE: Unsupervised Feature Learning for 3D Point Clouds from Multiple Angles by Joint Self-Reconstruction and Half-to-Half Prediction	-
Point-M2AE	92.9	Point-M2AE: Multi-scale Masked Autoencoders for Hierarchical Point Cloud Pre-training
FoldingNet	88.4	FoldingNet: Point Cloud Auto-encoder via Deep Grid Deformation
PointOE	90.7	Self-supervised Learning of Point Clouds via Orientation Estimation
Point-Jigsaw	90.6	Self-Supervised Deep Learning on Point Clouds by Reconstructing Space	-
MID-FC	90.3	Unsupervised 3D Learning for Shape Analysis via Multiresolution Instance Discrimination
SO-Net	87.5	SO-Net: Self-Organizing Network for Point Cloud Analysis
STRL	90.9	Spatio-temporal Self-Supervised Representation Learning for 3D Point Clouds
PSG-Net	90.9	Progressive Seed Generation Auto-encoder for Unsupervised Point Cloud Learning	-
3D-GAN	83.3	Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling
Point-JEPA	93.7±0.2	Point-JEPA: A Joint Embedding Predictive Architecture for Self-Supervised Learning on Point Cloud	-
ReCon++	93.6	ShapeLLM: Universal 3D Object Understanding for Embodied Interaction
VIP-GAN	90.2	View Inter-Prediction GAN: Unsupervised Representation Learning for 3D Shapes by Learning Global Shape Memories to Support Local View Predictions	-
CrossMoCo	91.49	CrossMoCo: Multi-modal Momentum Contrastive Learning for Point Cloud

0 of 20 row(s) selected.