thusiyuan / cooperative_scene_parsing

Code for NeurIPS 2018: Cooperative Holisctic Scene Understanding: Unifying 3D Object, Layout, and Camera Pose Estimation
http://siyuanhuang.com/cooperative_parsing/main.html
MIT License
101 stars 19 forks source link

Are ground-truth files in metadata preprocessed #7

Closed shubham-goel closed 5 years ago

shubham-goel commented 5 years ago

Hi!

Most of the code uses mat/pickle files present in the metadata/sunrgbd/ folder as ground truth. A few examples of the files being used are:

Are they simply a reorganization of the ground truth data (SUNRGBDMeta2DBB_v2.mat, SUNRGBDMeta3DBB_v2.mat, etc.) provided by the SUNRGBD dataset, or have you performed any additional processing?

Thanks, Shubham

thusiyuan commented 5 years ago

Hi,

Those are simply the reorganization of the original mat file except the 3D layout. The original layouts are 3D polygons, we use the Manhattan World assumption to process them as 3D cuboid.

Best, Siyuan