RGB-D Datasets

RGB-D Datasets

Name	Depth Image	Color Image	Ground Truth
3DMatch	X	X	camera pose
TUM

3DMatch

• paper: 3DMatch: Learning Local Geometric Descriptors from RGB-D Reconstructions, CVPR 2017

• website: https://3dmatch.cs.princeton.edu/

• notes: deep learning

UW

• paper: A Large-Scale Hierarchical Multi-View RGB-D Object Dataset, ICRA 2011

• website: http://rgbd-dataset.cs.washington.edu/dataset.html

TUM

• website: https://vision.in.tum.de/data/datasets/rgbd-dataset/download

OpenLORIS

• weblink: https://shimo.im/docs/HhJj6XHYhdRQ6jjk/read

Stanford Bunny

• website: http://graphics.stanford.edu/data/3Dscanrep/

Berkley Angle

• website: http://graphics.berkeley.edu/resources/ScannedModels/index.html

RGB-D Reconstruction Datasets organized by 3DMatch

Name	Depth Image	Color Image	Ground Truth Pose	Calibration
SUN3D	X	X	X	Yes
7-Scenes	X	X	X	No
RGB-D Scenes v2	X	X	X	Yes
BundleFusion	X	X	X	?
Analysis by Synthesis	X		X	?

Dataset Format

Each scene is a folder containing one or more RGB-D video sequences. The folder contents are as follows:

• camera-intrinsics.txt - a text file with depth camera intrinsics (3x3 matrix in homogeneous coordinates)

• seq-XX

  • frame-XXXXXX.color.png - a 24-bit PNG RGB color image

  • frame-XXXXXX.depth.png - a 16-bit PNG depth image, aligned to its corresponding color image. Depth is saved in millimeters (mm). Invalid depth is set to 0.

  • frame-XXXXXX.pose.txt - a text file with the camera pose of the frame (camera-to-world, 4x4 matrix in homogeneous coordinates and in meters)