ScanNet-200

ScanNet-200 is a dataset used to evaluate 3D panoptic segmentation performance in conjunction with 3D reconstruction models.

Papers using ScanNet-200 (3)

Imgeonet: Image-induced Geometry-aware Voxel Representation For Multi-view 3D Object Detection2023 · 13 citesMSTA3D: Multi-scale Twin-attention For 3D Instance Segmentation2024 · 5 citesPano3D: Unified 3D Reconstruction and Panoptic Segmentation2026