SUN RGB-D

Introduced by Song et al. in SUN RGB-D: A RGB-D Scene Understanding Benchmark Suite

The SUN RGBD dataset contains 10335 real RGB-D images of room scenes. Each RGB image has a corresponding depth and segmentation map. As many as 700 object categories are labeled. The training and testing sets contain 5285 and 5050 images, respectively.

Source: Mix and match networks: multi-domain alignment for unpaired image-to-image translation

Homepage

Benchmarks

Add a new result Link an existing benchmark

Task	Dataset Variant	Best Model
Semantic Segmentation	SUN-RGBD	TokenFusion
3D Object Detection	SUN-RGBD val	Point-GCC+TR3D+FF
3D Object Detection	SUN-RGBD	CAGroup3D
Monocular 3D Object Detection	SUN RGB-D	IM3D
Room Layout Estimation	SUN RGB-D	IM3D
Object Detection In Indoor Scenes	SUN RGB-D	YONOD + CPPM
Scene Segmentation	SUN-RGBD	ICM
Robust Semi-Supervised RGBD Semantic Segmentation	SUN RGB-D	M3L
Monocular Depth Estimation	SUN-RGBD	RPSF
Scene Recognition	SUN-RGBD	OMNIVORE
Scene Classification (unified classes)	SUN-RGBD	EMSANet
Object Detection	SUN-RGBD val	CDSSD
Panoptic Segmentation	SUN-RGBD	EMSANet