Few Shot 3D Point Cloud Classification On 3

评估指标

Overall Accuracy
Standard Deviation

评测结果

各个模型在此基准测试上的表现结果

Paper TitleRepository
Point-JEPA95.03.6Point-JEPA: A Joint Embedding Predictive Architecture for Self-Supervised Learning on Point Cloud
ReCon++94.54.1ShapeLLM: Universal 3D Object Understanding for Embodied Interaction
3D-JEPA94.33.63D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning-
PointGPT94.33.3PointGPT: Auto-regressively Generative Pre-training from Point Clouds
Point-FEMAE94.0-Towards Compact 3D Representations via Point Feature Enhancement Masked Autoencoders
point2vec93.94.1Point2Vec for Self-Supervised Representation Learning on Point Clouds
PCP-MAE93.53.7PCP-MAE: Learning to Predict Centers for Point Masked Autoencoders
Point-RAE93.34.0Regress Before Construct: Regress Autoencoder for Point Cloud Self-supervised Learning
ReCon93.33.9Contrast with Reconstruct: Contrastive 3D Representation Learning Guided by Generative Pretraining
ACT93.34.0Autoencoders as Cross-Modal Teachers: Can Pretrained 2D Image Transformers Help 3D Representation Learning?
OTMae3D93.23.4--
IDPT92.8-Instance-aware Dynamic Prompt Tuning for Pre-trained Point Cloud Models
Point-LGMask92.64.3Point-LGMask: Local and Global Contexts Embedding for Point Cloud Pre-training with Multi-Ratio Masking-
Point-MAE92.64.1Masked Autoencoders for Point Cloud Self-supervised Learning
I2P-MAE92.65.0Learning 3D Representations from 2D Pre-trained Models via Image-to-Point Masked Autoencoders
Point-M2AE92.34.5Point-M2AE: Multi-scale Masked Autoencoders for Hierarchical Point Cloud Pre-training
MaskPoint91.44.0Masked Discrimination for Self-Supervised Learning on Point Clouds
Point-BERT91.05.4Point-BERT: Pre-training 3D Point Cloud Transformers with Masked Point Modeling
CrossMoCo88.73.9CrossMoCo: Multi-modal Momentum Contrastive Learning for Point Cloud-
OcCo+PointNet83.91.8Unsupervised Point Cloud Pre-Training via Occlusion Completion
0 of 31 row(s) selected.
Few Shot 3D Point Cloud Classification On 3 | SOTA | HyperAI超神经