Where is supplement mentioned in the paper about mseg-api HOT 4 CLOSED

Hi @steven5401 , thanks for your interest in our work. The supplement will be included in an upcoming expanded Arxiv preprint. Here's a table that should address your question:

Let us know if you have further questions.

from mseg-api.

Thank you very much. Why didn't you choose scannet as training set, but test set?

from mseg-api.

No problem. We had a large number of concerns about the quality of the 2d labels in ScanNet. As you know, ScanNet is also annotated in 3d, not 2d. Unlike Matterport3d and Stanford-2d-3d, ScanNet does have a 2d label benchmark. Because of its noisy nature, we didn't want to train on it, but it is a good proxy for performance in the real-world.

Here are a few examples from ScanNet:

Book vs. bookshelf seems arbitrary:

The non-white pixels in the center should be "Wall", but include "blinds"

The door label extends onto the glass window:

Bed and backpack boundaries are poorly aligned:

Bicycle extent is quite incorrect:

from mseg-api.

I understand it. Thank you for the prompt reply.

from mseg-api.