Semantic Segmentation On Scannetv2

Metrics

Mean IoU

Results

Performance results of various models on this benchmark

		Paper Title
CMX	61.3%	CMX: Cross-Modal Fusion for RGB-X Semantic Segmentation with Transformers
EMSANet (2x ResNet-34 NBt1D, PanopticNDT version)	60.0%	PanopticNDT: Efficient and Robust Panoptic Mapping
RFBNet	59.2%	RFBNet: Deep Multimodal Networks with Residual Fusion Blocks for RGB-D Semantic Segmentation
SSMA	57.7	Self-Supervised Model Adaptation for Multimodal Semantic Segmentation
EMSAFormer	56.4%	Efficient Multi-Task Scene Analysis with RGB-D Transformers
AdapNet++	50.3	Self-Supervised Model Adaptation for Multimodal Semantic Segmentation
3DMV (2d proj)	49.8%	3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation
MSeg1080_RVC	48.5%	MSeg: A Composite Dataset for Multi-domain Semantic Segmentation
PSPNet	47.5%	Pyramid Scene Parsing Network
ENet	37.6%	ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation
ScanNet (2d proj)	33.0%	ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes
Floors are Flat	-	Floors are Flat: Leveraging Semantics for Real-Time Surface Normal Prediction

0 of 12 row(s) selected.