Home News Papers Tutorials Datasets Wiki SOTA LLM Models GPU Leaderboard Events

English

3D Semantic Segmentation On Scannet200

Metrics

test mIoU

val mIoU

Results

Performance results of various models on this benchmark

Model Name	test mIoU	val mIoU	Paper Title	Repository
DITR	44.9	41.2	DINO in the Room: Leveraging 2D Foundation Models for 3D Segmentation
PPT+SparseUNet	33.2	31.9	Towards Large-scale 3D Representation Learning with Multi-dataset Point Prompt Training
Pamba	37.1	36.3	Pamba: Enhancing Global Interaction in Point Clouds via State Space Model	-
OctFormer	32.5	32.6	OctFormer: Octree-based Transformers for 3D Point Clouds
PonderV2 + SparseUNet	34.6	32.3	PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm
OneFormer3D	-	30.1	OneFormer3D: One Transformer for Unified Point Cloud Segmentation
OA-CNNs	32.3	33.3	OA-CNNs: Omni-Adaptive Sparse CNNs for 3D Semantic Segmentation
LSK3DNet	-	33.1	LSK3DNet: Towards Effective and Efficient 3D Perception with Large Sparse Kernels
CSC	24.9	26.4	Exploring Data-Efficient 3D Scene Understanding with Contrastive Scene Contexts
MinkUNet	25.3	25.0	4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks
ODIN	36.8	40.5	ODIN: A Single Model for 2D and 3D Segmentation
Sonata + PTv3	-	36.8	Sonata: Self-Supervised Learning of Reliable Point Representations
BFANet	36.0	37.3	BFANet: Revisiting 3D Semantic Segmentation with Boundary Feature Analysis
PTv3 + PPT	39.3	36.0	Point Transformer V3: Simpler, Faster, Stronger
LGround	27.2	28.8	Language-Grounded Indoor 3D Semantic Segmentation in the Wild
PTv3 ArKitLabelmaker	41.4	40.3	ARKit LabelMaker: A New Scale for Indoor 3D Scene Understanding

0 of 16 row(s) selected.