HyperAI초신경

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

한국어

HyperAI초신경

Point Cloud Registration On Kitti Fcgf

평가 지표

RE (all)

Recall (0.6m, 5 degrees)

TE (all)

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	RE (all)	Recall (0.6m, 5 degrees)	TE (all)	Paper Title	Repository
PCAM-Sparse + ICP	1.04	97.4	0.17	PCAM: Product of Cross-Attention Matrices for Rigid Registration of Point Clouds
FGR	-	0.2	-	Fast Global Registration
FCGF	-	98.2	-	Fully Convolutional Geometric Features
PCAM-soft + ICP	0.79	98	0.12	PCAM: Product of Cross-Attention Matrices for Rigid Registration of Point Clouds
PCAM-Sparse	1.17	96.5	0.22	PCAM: Product of Cross-Attention Matrices for Rigid Registration of Point Clouds
Exhaustive Grid Search	-	94.59	-	Addressing the generalization of 3D registration methods with a featureless baseline and an unbiased benchmark
RANSAC	-	34.2	-	Fast Point Feature Histograms (FPFH) for 3D Registration
DGR (RE (all), TE(all) are reported in PCAM)	1.62	96.9	0.34	Deep Global Registration
PCAM - Soft	1.00	97.2	0.18	PCAM: Product of Cross-Attention Matrices for Rigid Registration of Point Clouds
GeoTransformer	-	99.5	-	Geometric Transformer for Fast and Robust Point Cloud Registration
DGR + ICP (RE (all), TE(all) are reported in PCAM)	1.43	98.2	0.16	Deep Global Registration

0 of 11 row(s) selected.