Stanford-2D-3DS

The 'Stanford-2D-3DS' dataset is a benchmark that contains panoramic images used to evaluate semantic segmentation models, particularly in the context of RGB, RGB-D, and RGB-D-N modalities.

Papers using Stanford-2D-3DS (2)

PanoSAMic: Panoramic Image Segmentation from SAM Feature Encoding and Dual View Fusion2026Power of Boundary and Reflection: Semantic Transparent Object Segmentation using Pyramid Vision Transformer with Transparent Cues2025