Rendered output inconsistent with provided dataset images

[Z-MU-Z]

Hi authors, thank you for the great work and for releasing the MesaTask dataset!

I’m currently using the provided layout files and GLB assets to render the scenes using vis_single.py.
However, I noticed that the rendered output is not consistent with the images shown in the dataset (e.g., the example renders on Hugging Face ). Specifically:

Rendered with vis_single.py	Official example

I followed the instructions and ran:

python vis_single.py MesaTask-10K/Layout_info/office_table/office_table_0001/layout.json --output_dir vis_data

[jinkun-hao]

Thank you for your attention to our project.
The discrepancy in rendered images is caused by issues with the 3D model assets.
Due to copyright restrictions, we are unable to provide some of the original 3D assets. For the office_table_0001 scene, the laptop asset can be downloaded from the following https://www.blenderkit.com/get-blenderkit/5adb5d73-fb75-4380-a607-cc7d451e7344/
Additionally, for all 3D model assets in Asset_annotation.json where the "source" : "gen", if you need to obtain the original 3D model assets, you can download them via the URL pattern:www.blenderkit.com/get-blenderkit/{uid}/(Note: Replace {uid} with the unique ID corresponding to each asset in the JSON file.)