Papers with Code · CC BY-SA 4.0

S3DIS Area5

아카이브 평가 61건 · 마지막 적재/변경 2026-10-30

평가 기록

과거 수집 기록이며 최신 순위가 아닙니다. 데이터셋 버전·split·실험 조건과 지표 방향이 검증되지 않은 결과는 직접 비교할 수 없습니다. 날짜는 논문 날짜이며 최신 평가일을 뜻하지 않습니다.

표가 화면보다 넓으면 좌우로 스크롤하여 논문과 구현 링크를 확인하세요.

원본 기록 순서 · 순위 미부여
모델	원본 지표·값	논문	구현
Sonata + PTv3	mIoU 76.0 oAcc 93.0 mAcc 81.6	Sonata: Self-Supervised Learning of Reliable Point Representations 2025-03-20	코드 1 · 코드 2
OmniVec	mIoU 75.9	OmniVec: Learning robust representations with cross modal sharing 2023-11-07	연결 없음
D-DITR	mIoU 75.0	DINO in the Room: Leveraging 2D Foundation Models for 3D Segmentation 2025-03-24	코드 1
PTv3 + PPT	mIoU 74.7 oAcc 92.0 mAcc 80.1	Point Transformer V3: Simpler, Faster, Stronger 2023-12-15	코드 1 · 코드 2 · 코드 3
Swin3D-L	mIoU 74.5 oAcc 92.7 mAcc 80.5 Number of params N/A	Swin3D: A Pretrained Transformer Backbone for 3D Indoor Scene Understanding 2023-04-14	코드 1 · 코드 2
DeLA	mIoU 74.1 oAcc 92.2 mAcc 80.0 Number of params 7.0M	Decoupled Local Aggregation for Point Cloud Learning 2023-08-31	코드 1
DITR	mIoU 74.1	DINO in the Room: Leveraging 2D Foundation Models for 3D Segmentation 2025-03-24	코드 1
Ours	mIoU 73.6 oAcc 93.0 mAcc 80.2	Beyond local patches: Preserving global–local interactions by enhancing self-attention via 3D point cloud tokenization 2024-11-01	연결 없음
ConDaFormer	mIoU 73.5 oAcc 92.4 mAcc 78.9	ConDaFormer: Disassembled Transformer with Local Structure Enhancement for 3D Point Cloud Understanding 2023-12-18	코드 1
LPFP(Point Transformer*)	mIoU 73.5 oAcc 92.0 mAcc 78.7 Number of params 31.2M FLOPs 24.2G	A Large-Scale Network Construction and Lightweighting Method for Point Cloud Semantic Segmentation 2024-03-21	코드 1
KPConvX-L	mIoU 73.5 oAcc 91.7 mAcc 78.7	KPConvX: Modernizing Kernel Point Convolution with Kernel Attention 2024-05-21	코드 1
Pamba	mIoU 73.5	Pamba: Enhancing Global Interaction in Point Clouds via State Space Model 2024-06-25	연결 없음
SPG(PTv2)	mIoU 73.3 oAcc 91.9 mAcc 79.5	Subspace Prototype Guidance for Mitigating Class Imbalance in Point Cloud Semantic Segmentation 2024-08-20	코드 1
PonderV2 + SparseUNet	mIoU 73.2 oAcc 92.2 mAcc 79.0	PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm 2023-10-12	코드 1
PointHR	mIoU 73.2 oAcc 91.8 mAcc 78.7	PointHR: Exploring High-Resolution Architectures for 3D Point Cloud Segmentation 2023-10-11	코드 1
PPT + SparseUNet	mIoU 72.7 oAcc 91.5 mAcc 78.2 Number of params N/A	Towards Large-scale 3D Representation Learning with Multi-dataset Point Prompt Training 2023-08-18	코드 1
SAT (FAT)	mIoU 72.6 mAcc 78.8 Number of params N/A	SAT: Size-Aware Transformer for 3D Point Cloud Semantic Segmentation 2023-01-17	연결 없음
PTv2	mIoU 72.6 oAcc 91.6 mAcc 78.0 Number of params N/A	Point Transformer V2: Grouped Vector Attention and Partition-based Pooling 2022-10-11	코드 1 · 코드 2
PT + ERDA	mIoU 72.6	연결 없음 논문 날짜 미기재	연결 없음
PointVector-XL	mIoU 72.3 oAcc 91 mAcc 78.1	PointVector: A Vector Representation In Point Cloud Analysis 2022-05-21	코드 1
WindowNorm+StratifiedTransformer	mIoU 72.2 oAcc 91.4 mAcc 78.2 Number of params N/A	Window Normalization: Enhancing Point Cloud Understanding by Unifying Inconsistent Point Densities 2022-12-05	코드 1
Diffusion Unit	mIoU 72.2 oAcc 91.3 mAcc 78.1 Number of params 8.1M	Diffusion Unit: Interpretable Edge Enhancement and Suppression Learning for 3D Point Cloud Segmentation 2022-09-20	코드 1
CSC+MinkUNet	mIoU 72.2	Exploring Data-Efficient 3D Scene Understanding with Contrastive Scene Contexts 2020-12-16	코드 1 · 코드 2
StratifiedTransformer	mIoU 72.0 oAcc 91.5 mAcc 78.1 Number of params 8.0M	Stratified Transformer for 3D Point Cloud Segmentation 2022-03-28	코드 1 · 코드 2 · 코드 3 · 코드 4
PointTransformer+CBL	mIoU 71.6 oAcc 91.2 mAcc 77.9 Number of params N/A	Contrastive Boundary Learning for Point Cloud Segmentation 2022-03-10	코드 1
WindowNorm+PointTransformer	mIoU 71.4 oAcc 91.1 mAcc 77.9 Number of params N/A	Window Normalization: Enhancing Point Cloud Understanding by Unifying Inconsistent Point Densities 2022-12-05	코드 1
PointMixer	mIoU 71.4 mAcc 77.4 Number of params 6.5M	PointMixer: MLP-Mixer for Point Cloud Understanding 2021-11-22	코드 1 · 코드 2 · 코드 3
PointMetaBase-XXL	mIoU 71.3±0.7 Number of params N/A	Meta Architecture for Point Cloud Analysis 2022-11-26	코드 1
PointNeXt	mIoU 71.1 oAcc 91.0 mAcc 77.2 Number of params 41.6M	PointNeXt: Revisiting PointNet++ with Improved Training and Scaling Strategies 2022-06-09	코드 1 · 코드 2 · 코드 3
SPoTr	mIoU 70.8 oAcc 90.7 mAcc 76.4 Number of params N/A	Self-positioning Point-based Transformer for Point Cloud Understanding 2023-03-29	코드 1