3D Point Cloud Linear Classification on ModelNet40

Metric: Overall Accuracy (higher is better)

LeaderboardDataset

Loading chart...

Results

Hide extra data

Sort:

#	Model↕	Overall Accuracy▼	Extra Data	Paper	Date↕	Code
1	ReCon++	93.6	No	ShapeLLM: Universal 3D Object Understanding for ...	2024-02-27	Code
2	I2P-MAE	93.4	Yes	Learning 3D Representations from 2D Pre-trained ...	2022-12-13	Code
3	ReCon	93.4	Yes	Contrast with Reconstruct: Contrastive 3D Repres...	2023-02-05	Code
4	Point-M2AE	92.9	Yes	Point-M2AE: Multi-scale Masked Autoencoders for ...	2022-05-28	Code
5	IAE (DGCNN)	92.1	Yes	Implicit Autoencoder for Point-Cloud Self-Superv...	2022-01-03	Code
6	AdaCrossNet	91.8	Yes	-	-	Code
7	CrossMoCo	91.49	No	-	-	Code
8	CrossPoint	91.2	Yes	CrossPoint: Self-Supervised Cross-Modal Contrast...	2022-03-01	Code
9	STRL	90.9	No	Spatio-temporal Self-Supervised Representation L...	2021-09-01	Code
10	PSG-Net	90.9	No	Progressive Seed Generation Auto-encoder for Uns...	2021-12-09	-
11	PointOE	90.7	No	Self-supervised Learning of Point Clouds via Ori...	2020-08-01	Code
12	Point-Jigsaw	90.6	No	Self-Supervised Deep Learning on Point Clouds by...	2019-01-24	-
13	MID-FC	90.3	No	Unsupervised 3D Learning for Shape Analysis via ...	2020-08-03	Code
14	VIP-GAN	90.2	No	View Inter-Prediction GAN: Unsupervised Represen...	2018-11-07	-
15	OcCo	89.2	No	Unsupervised Point Cloud Pre-Training via Occlus...	2020-10-02	Code
16	FoldingNet	88.4	No	FoldingNet: Point Cloud Auto-encoder via Deep Gr...	2017-12-19	Code
17	MAE-VAE	88.4	No	Multi-Angle Point Cloud-VAE: Unsupervised Featur...	2019-07-30	-
18	SO-Net	87.5	No	SO-Net: Self-Organizing Network for Point Cloud ...	2018-03-12	Code
19	3D-GAN	83.3	No	Learning a Probabilistic Latent Space of Object ...	2016-10-24	Code

#1ReCon++SOTA
93.6
Overall Accuracy· 2024-02-27
ShapeLLM: Universal 3D Object Understanding for Embodied Interaction Code
#2I2P-MAESOTA
93.4
Overall Accuracy· Extra Data· 2022-12-13
Learning 3D Representations from 2D Pre-trained Models via Image-to-Point Masked Autoencoders Code
#3ReCon
93.4
Overall Accuracy· Extra Data· 2023-02-05
Contrast with Reconstruct: Contrastive 3D Representation Learning Guided by Generative Pretraining Code
#4Point-M2AESOTA
92.9
Overall Accuracy· Extra Data· 2022-05-28
Point-M2AE: Multi-scale Masked Autoencoders for Hierarchical Point Cloud Pre-training Code
#5IAE (DGCNN)SOTA
92.1
Overall Accuracy· Extra Data· 2022-01-03
Implicit Autoencoder for Point-Cloud Self-Supervised Representation Learning Code
#6AdaCrossNet
91.8
Overall Accuracy· Extra Data
No paperCode
#7CrossMoCo
91.49
Overall Accuracy
No paperCode
#8CrossPoint
91.2
Overall Accuracy· Extra Data· 2022-03-01
CrossPoint: Self-Supervised Cross-Modal Contrastive Learning for 3D Point Cloud Understanding Code
#9STRLSOTA
90.9
Overall Accuracy· 2021-09-01
Spatio-temporal Self-Supervised Representation Learning for 3D Point Clouds Code
#10PSG-Net
90.9
Overall Accuracy· 2021-12-09
Progressive Seed Generation Auto-encoder for Unsupervised Point Cloud Learning
#11PointOESOTA
90.7
Overall Accuracy· 2020-08-01
Self-supervised Learning of Point Clouds via Orientation Estimation Code
#12Point-JigsawSOTA
90.6
Overall Accuracy· 2019-01-24
Self-Supervised Deep Learning on Point Clouds by Reconstructing Space
#13MID-FC
90.3
Overall Accuracy· 2020-08-03
Unsupervised 3D Learning for Shape Analysis via Multiresolution Instance Discrimination Code
#14VIP-GANSOTA
90.2
Overall Accuracy· 2018-11-07
View Inter-Prediction GAN: Unsupervised Representation Learning for 3D Shapes by Learning Global Shape Memories to Support Local View Predictions
#15OcCo
89.2
Overall Accuracy· 2020-10-02
Unsupervised Point Cloud Pre-Training via Occlusion Completion Code
#16FoldingNetSOTA
88.4
Overall Accuracy· 2017-12-19
FoldingNet: Point Cloud Auto-encoder via Deep Grid Deformation Code
#17MAE-VAE
88.4
Overall Accuracy· 2019-07-30
Multi-Angle Point Cloud-VAE: Unsupervised Feature Learning for 3D Point Clouds from Multiple Angles by Joint Self-Reconstruction and Half-to-Half Prediction
#18SO-Net
87.5
Overall Accuracy· 2018-03-12
SO-Net: Self-Organizing Network for Point Cloud Analysis Code
#193D-GANSOTA
83.3
Overall Accuracy· 2016-10-24
Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling Code