This article is an original article of the blogger, and may not be reproduced without the permission of the blogger.
This article is a series of articles in the column "python 3D point cloud from basics to deep learning", and the address is "https://blog.csdn.net/suiyingy/article/details/124017716".
SUN RGB-D is a data set about indoor scene understanding released by Princeton University. It contains a total of 10335 samples, of which the number of training samples and verification test samples are 5285 and 5050, respectively. Each sample contains color image (RGB) and depth (D) information, and is annotated in 2D and 3D, respectively. The specific details of the dataset will be supplemented later. Here we mainly introduce how to use the SUN RGB-D dataset in mmdection3d.
1 SUN RGB-D dataset download
The download address of the SUN RGB-D dataset is https://rgbd.cs.princeton.edu/data/, mainly downloading files such as SUNRGBD.zip, SUNRGBDMeta2DBB_v2.mat, SUNRGBDMeta3DBB_v2.mat and SUNRGBDtoolbox.zip.
Figure 1 SUN RGB-D dataset download list
The downloaded SUNRGBD.zip, SUNR