96lives/ssen

Code for Spatial Semantic Embedding Network:Fast 3D Instance Segmentation with Deep Metric Learning

ssen's Issues

License

First of all thanks for providing the awesome code.

Under which license is it being published? I.g. is it allowed to use it (or parts of it) for research, as well as commercial products?

Does the last line "test" wrong in scannet_val.txt?

there is no scene named "test"

Question about way to train instant segmentation model

First of all, thank you for sharing the nice code! I have a few questions ...

• In the paper, it is specified that the semantic model requires pre-trained MinkowskiNet.
  However, for instant segmentation model, I'm not so clear whether I have to use pre-trained weight or not.
  Training an instant segmentation model from scratch is enough to reproduce the result? Which setting did you use in your paper?

• In code, It seems like SSEN load pre-trained mikowskiNet weight at the start of training. However, I found out that loading weight files using init() code of SemanticSegModel or InstantSegModel class does not work properly (Maybe strict=False could be the reason since preprocessing_semnatic.py works well.).

• Do you have any plan to release the evaluation code(Which compute mAP) for validation split?

How to visualize the instance segmentation results?

Hello. Thank you for your nice work.
I have a question. How to use the trained model to visualize the results? The trained model is different the instance segmentation model that you upload to google drive.
Hope your reply. Thank you.