Skip to content

Latest commit

 

History

History
10 lines (6 loc) · 544 Bytes

README.md

File metadata and controls

10 lines (6 loc) · 544 Bytes

point_recalibration

Implementation of the re-calibration blocks introduced in the article "Recalibration of Neural Networks for Point Cloud Analysis" presented at 3DV 2020.

Authors: Ignacio Sarasua; Sebastian Pölsterl; Christian Wachinger

rec_blocks.py : includes the implementation of the different blocks

dgcnn_crb.py: includes an example of the integration of the channel re-calibration block in DGCNN original's implementation (https://github.com/AnTao97/dgcnn.pytorch). Check their repository for detais on how to run their model.