HyperAI

Few Shot 3D Point Cloud Classification On 2

Métriques

Overall Accuracy
Standard Deviation

Résultats

Résultats de performance de divers modèles sur ce benchmark

Nom du modèle
Overall Accuracy
Standard Deviation
Paper TitleRepository
point2vec98.71.2Point2Vec for Self-Supervised Representation Learning on Point Clouds
Point-JEPA99.20.8Point-JEPA: A Joint Embedding Predictive Architecture for Self-Supervised Learning on Point Cloud-
OcCo+PointNet92.41.6Unsupervised Point Cloud Pre-Training via Occlusion Completion
IDPT97.9-Instance-aware Dynamic Prompt Tuning for Pre-trained Point Cloud Models
PointNet++42.3914.2PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space
PCP-MAE99.10.8PCP-MAE: Learning to Predict Centers for Point Masked Autoencoders
PointCNN68.647.0PointCNN: Convolution On X-Transformed Points
Point-RAE98.71.3Regress Before Construct: Regress Autoencoder for Point Cloud Self-supervised Learning
ACT98.01.4Autoencoders as Cross-Modal Teachers: Can Pretrained 2D Image Transformers Help 3D Representation Learning?
GPr-Net + Hyp (1024)82.00.9GPr-Net: Geometric Prototypical Network for Point Cloud Few-Shot Learning
GPr-Net + Hyp (512)82.71.3GPr-Net: Geometric Prototypical Network for Point Cloud Few-Shot Learning
PointNet57.8115.5PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
Point-LGMask98.11.4Point-LGMask: Local and Global Contexts Embedding for Point Cloud Pre-training with Multi-Ratio Masking
Point-MAE97.81.8Masked Autoencoders for Point Cloud Self-supervised Learning
OcCo+DGCNN92.51.9Unsupervised Point Cloud Pre-Training via Occlusion Completion
3D-JEPA98.80.43D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning-
MaskPoint97.21.7Masked Discrimination for Self-Supervised Learning on Point Clouds
SSFSL+PointNet68.909.4Self-Supervised Few-Shot Learning on Point Clouds
DGCNN40.814.6Dynamic Graph CNN for Learning on Point Clouds
GPr-Net + Euc (1024)75.12.1GPr-Net: Geometric Prototypical Network for Point Cloud Few-Shot Learning
0 of 30 row(s) selected.