3D Point Cloud Linear Classification On

Overall Accuracy

评测结果

各个模型在此基准测试上的表现结果

比较表格

模型名称	Overall Accuracy
implicit-autoencoder-for-point-cloud-self	92.1
learning-3d-representations-from-2d-pre	93.4
adacrossnet-adaptive-dynamic-loss-weighting	91.4
contrast-with-reconstruct-contrastive-3d	93.4
pre-training-by-completing-point-clouds	89.2
crosspoint-self-supervised-cross-modal	91.2
multi-angle-point-cloud-vae-unsupervised	88.4
point-m2ae-multi-scale-masked-autoencoders	92.9
foldingnet-point-cloud-auto-encoder-via-deep	88.4
self-supervised-learning-of-point-clouds-via	90.7
context-prediction-for-unsupervised-deep	90.6
unsupervised-3d-learning-for-shape-analysis	90.3
so-net-self-organizing-network-for-point	87.5
spatio-temporal-self-supervised	90.9
progressive-seed-generation-auto-encoder-for-1	90.9
learning-a-probabilistic-latent-space-of	83.3
point-jepa-a-joint-embedding-predictive	93.7±0.2
shapellm-universal-3d-object-understanding	93.6
view-inter-prediction-gan-unsupervised	90.2
crossmoco-multi-modal-momentum-contrastive	91.49