Shape Representation Of 3D Point Clouds on ModelNet40

Metric: Accuracy (%) (higher is better)

LeaderboardDataset

Loading chart...

Results

Hide extra data

Sort:

#	Model↕	Accuracy (%)▼	Extra Data	Paper	Date↕	Code
1	Uni3D	88.2	Yes	Uni3D: Exploring Unified 3D Representation at Sc...	2023-10-10	Code
2	ViT-Lens	87.6	Yes	ViT-Lens: Initiating Omni-Modal Exploration thro...	2023-08-20	Code
3	ReCon++	87.3	Yes	ShapeLLM: Universal 3D Object Understanding for ...	2024-02-27	Code
4	MixCon3D-PointBERT	86.8	Yes	Sculpting Holistic 3D Representation in Contrast...	2023-11-03	Code
5	TAMM-PointBERT (+dlign)	86.2	Yes	OpenDlign: Open-World Point Cloud Understanding ...	2024-04-25	Code
6	OpenShape-PointBERT (+dlign)	85.4	Yes	OpenDlign: Open-World Point Cloud Understanding ...	2024-04-25	Code
7	OpenShape-PointBERT	85.3	Yes	OpenShape: Scaling Up 3D Shape Representation To...	2023-05-18	Code
8	OpenShape-SparseConv (+dlign)	85	Yes	OpenDlign: Open-World Point Cloud Understanding ...	2024-04-25	Code
9	OpenShape-SparseConv	83.4	Yes	OpenShape: Scaling Up 3D Shape Representation To...	2023-05-18	Code
10	OpenDlign	82.6	Yes	OpenDlign: Open-World Point Cloud Understanding ...	2024-04-25	Code
11	PointCLIP V2	64.22	Yes	PointCLIP V2: Prompting CLIP and GPT for Powerfu...	2022-11-21	Code
12	ReCon	61.7	Yes	Contrast with Reconstruct: Contrastive 3D Repres...	2023-02-05	Code
13	ULIP + PointMLP	61.5	Yes	ULIP: Learning a Unified Representation of Langu...	2022-12-10	Code
14	ULIP + PointBERT	60.4	Yes	ULIP: Learning a Unified Representation of Langu...	2022-12-10	Code
15	CLIP2Point	49.38	Yes	CLIP2Point: Transfer CLIP to Point Cloud Classif...	2022-10-03	Code
16	PointCLIP	20.18	Yes	PointCLIP: Point Cloud Understanding by CLIP	2021-12-04	Code

#1Uni3DSOTA
88.2
Accuracy (%)· Extra Data· 2023-10-10
Uni3D: Exploring Unified 3D Representation at Scale Code
#2ViT-LensSOTA
87.6
Accuracy (%)· Extra Data· 2023-08-20
ViT-Lens: Initiating Omni-Modal Exploration through 3D Insights Code
#3ReCon++
87.3
Accuracy (%)· Extra Data· 2024-02-27
ShapeLLM: Universal 3D Object Understanding for Embodied Interaction Code
#4MixCon3D-PointBERT
86.8
Accuracy (%)· Extra Data· 2023-11-03
Sculpting Holistic 3D Representation in Contrastive Language-Image-3D Pre-training Code
#5TAMM-PointBERT (+dlign)
86.2
Accuracy (%)· Extra Data· 2024-04-25
OpenDlign: Open-World Point Cloud Understanding with Depth-Aligned Images Code
#6OpenShape-PointBERT (+dlign)
85.4
Accuracy (%)· Extra Data· 2024-04-25
OpenDlign: Open-World Point Cloud Understanding with Depth-Aligned Images Code
#7OpenShape-PointBERTSOTA
85.3
Accuracy (%)· Extra Data· 2023-05-18
OpenShape: Scaling Up 3D Shape Representation Towards Open-World Understanding Code
#8OpenShape-SparseConv (+dlign)
85
Accuracy (%)· Extra Data· 2024-04-25
OpenDlign: Open-World Point Cloud Understanding with Depth-Aligned Images Code
#9OpenShape-SparseConv
83.4
Accuracy (%)· Extra Data· 2023-05-18
OpenShape: Scaling Up 3D Shape Representation Towards Open-World Understanding Code
#10OpenDlign
82.6
Accuracy (%)· Extra Data· 2024-04-25
OpenDlign: Open-World Point Cloud Understanding with Depth-Aligned Images Code
#11PointCLIP V2SOTA
64.22
Accuracy (%)· Extra Data· 2022-11-21
PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning Code
#12ReCon
61.7
Accuracy (%)· Extra Data· 2023-02-05
Contrast with Reconstruct: Contrastive 3D Representation Learning Guided by Generative Pretraining Code
#13ULIP + PointMLP
61.5
Accuracy (%)· Extra Data· 2022-12-10
ULIP: Learning a Unified Representation of Language, Images, and Point Clouds for 3D Understanding Code
#14ULIP + PointBERT
60.4
Accuracy (%)· Extra Data· 2022-12-10
ULIP: Learning a Unified Representation of Language, Images, and Point Clouds for 3D Understanding Code
#15CLIP2PointSOTA
49.38
Accuracy (%)· Extra Data· 2022-10-03
CLIP2Point: Transfer CLIP to Point Cloud Classification with Image-Depth Pre-training Code
#16PointCLIPSOTA
20.18
Accuracy (%)· Extra Data· 2021-12-04
PointCLIP: Point Cloud Understanding by CLIP Code