search

liuxinhai / Point2Sequence

Point2Sequence: Learning the Shape Representation of 3D Point Clouds with an Attention-based Sequence to Sequence Network
Other
38 stars 7 forks source link

about s3dis and scannet #4

Open longmalongma opened 4 years ago

longmalongma commented 4 years ago

why do not you try to use your model to implement task of scene segmentation on s3dis of scannet dataset?can your model make it?

liuxinhai commented 4 years ago

I think the answer is yes. With some modifications, our model can be applied to large scale point cloud data including s3dis dataset and scannet dataset. Due to limited time before， we have never tried on these datasets.