jerrypiglet / rui-indoorinv-data Goto Github PK

Hi thanks for your great work!
I tried to download the data and reproduce the data processing pipeline for real captures.
However I found that the raw_images here do not match with the processed images.
Here is one of the example:
scene used in the paper:

RAW images provided:

We can see that the chairs arrangement and objects on the table look different
Is it possible to get access to the RAW images of conference room that are used in the paper?
Thanks a lot for your help!

Hi, I ran the cmd python load_mitsubaScene3D.py --scene kitchen_mi --render_2d --renderer blender and got the transform.json file. However, the generated pose looks like this compared to the original pose(openrooms-txt or blender-npy). I think this can be fixed by a coordinate conversion like y-up to z-up when dumping the pose.

Hi,

It seems that this file is missing from the current branch, which is used in sample_poses().

Thanks

Hi,thanks for your share! I have run some aformentioned command in README_indoor_synthetic.md. However, when i run the command "python load_mitsubaScene3D.py --scene kitchen_mi --render_2d --renderer mi", it has an error that i don't have transforms.json file. I don't know where is the problem.