Our new X account is live! Follow @wizwand_team for updates

Few-shot object classification on ModelNet40 (test)

96.3Accuracy (5-way, 10-shot)

Point-MAE

Updated 4d ago

Evaluation Results

Method	Links
Point-MAE 2022.03		96.3	97.8	92.6	95
MaskPoint 2022.03		95	97.2	91.4	93.4
Point-BERT 2022.03		94.6	96.3	91	92.7
Point-BERT 2022.03		94.6	96.3	91	92.7
Transformer-OcCo 2022.03		94	95.9	89.4	92.4
PointViT-OcCo 2022.03		94	95.9	89.4	92.4
DGCNN + CrossPoint 2022.03		92.5	94.9	83.6	87.9
DGCNN-OcCo 2022.03		91.9	93.9	86.4	91.3
DGCNN 2022.03		91.8	93.4	86.3	90.9
PointNet + CrossPoint 2022.03		90.9	93.5	84.6	90.2
DGCNN + OcCo 2022.03		90.6	92.5	82.9	86.5
DGCNN-OcCo 2022.03		90.6	92.5	82.9	86.5
PointNet + OcCo 2022.03		89.7	92.4	83.9	89.7
Transformer-rand 2022.03		87.8	93.3	84.6	89.4
PointViT 2022.03		87.8	93.3	84.6	89.4
PointNet + Jigsaw 2022.03		66.5	69.2	56.9	66.5
PointCNN 2022.03		65.4	68.6	46.6	50
RSCNN 2022.03		65.4	68.6	46.6	50
PointNet + cTree 2022.03		63.2	68.9	49.2	50.1
DGCNN + cTree 2022.03		60	65.7	48.5	53
3D-GAN 2022.03		55.8	65.8	40.3	48.4
PointNet + Rand 2022.03		52	57.8	46.6	35.2
3D-PointCapsNet 2022.03		42.3	53	38	27.2
Latent-GAN 2022.03		41.6	46.2	32.9	25.5
PointNet++ 2022.03		38.5	42.4	23.1	18.8
DGCNN + Jigsaw 2022.03		34.3	42.2	26	29.9
FoldingNet 2022.03		33.4	35.8	18.6	15.4
DGCNN + Rand 2022.03		31.6	40.8	19.9	16.9
DGCNN-rand 2022.03		31.6	40.8	19.9	16.9