search

drprojects / superpoint_transformer

Official PyTorch implementation of Superpoint Transformer introduced in [ICCV'23] "Efficient 3D Semantic Segmentation with Superpoint Transformer" and SuperCluster introduced in [3DV'24 Oral] "Scalable 3D Panoptic Segmentation As Superpoint Graph Clustering"
MIT License
560 stars 72 forks source link

how to extracting the features? #10

Closed aksahveoffice closed 1 year ago

aksahveoffice commented 1 year ago

i am not getting feature extraction code.

drprojects commented 1 year ago

Hi, please provide a more detailed explanation of the issue you are encountering.

drprojects commented 1 year ago

Taking a wild guess: I am assuming you would like to know where to find the extracted superpoint features. If so, a brief investigation of the code should have lead you here. The output of self.net are the extracted features, before calling self.head on them to produce semantic segmentation logits.

NB: The above-mentionned features are superpoint-wise features and not point-wsie features. For more details on that your can read our paper and check out the code documentation and comments.

Without further details from you, I am closing this issue.