HyperAI초신경

홈 뉴스 최신 연구 논문 튜토리얼 데이터셋 백과사전 SOTA LLM 모델 GPU 랭킹 컨퍼런스

한국어

HyperAI초신경

Point Cloud Registration On 3Dlomatch 10 30

평가 지표

Recall ( correspondence RMSE below 0.2)

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	Recall ( correspondence RMSE below 0.2)	Paper Title	Repository
REGTR	64.8	REGTR: End-to-end Point Cloud Correspondences with Transformers	-
3DSN (reported in PREDATOR)	33	The Perfect Match: 3D Point Cloud Matching with Smoothed Densities	-
PCAM (reported in REGTR)	54.9	PCAM: Product of Cross-Attention Matrices for Rigid Registration of Point Clouds	-
DGR (reported in REGTR)	48.7	Deep Global Registration	-
FCGF (reported in PREDATOR)	40.1	Fully Convolutional Geometric Features
D3Feat (reported in PREDATOR)	37.2	D3Feat: Joint Learning of Dense Detection and Description of 3D Local Features	-
Predator-1k	62.5	PREDATOR: Registration of 3D Point Clouds with Low Overlap	-
OMNet (reported in REGTR)	8.4	OMNet: Learning Overlapping Mask for Partial-to-Partial Point Cloud Registration	-
NgeNet	71.9	Leveraging Inlier Correspondences Proportion for Point Cloud Registration	-
GeoTransformer - P2PNet	74	Geometric Transformer for Fast and Robust Point Cloud Registration	-
Predator-5k	59.8	PREDATOR: Registration of 3D Point Clouds with Low Overlap	-
Predator-NR	24	PREDATOR: Registration of 3D Point Clouds with Low Overlap	-

0 of 12 row(s) selected.

Point Cloud Registration On 3Dlomatch 10 30 | SOTA | HyperAI초신경