Paper Image

Detecting 3D patterns regardless of orientation

Published on:

28 March 2024

Primary Category:

Computer Vision and Pattern Recognition

Paper Authors:

Dmitrii Zhemchuzhnikov,

Sergei Grudinin

Bullets

Key Details

Presents ILPO-Net, detecting arbitrary 3D patterns invariant to orientation

Integrates new convolution operator based on Wigner matrices

Reduces parameters needed, outperforms baselines on volumetric data

Invariance enables applications across disciplines

AI generated summary

Detecting 3D patterns regardless of orientation

This paper introduces ILPO-Net, a novel deep learning technique to recognize 3D patterns in volumetric data, with built-in invariance to the patterns' orientations. When tested on biomedical and protein structure datasets, it achieved higher accuracy than previous methods while using far fewer parameters.

Answers from this paper

Comments

No comments yet, be the first to start the conversation...

Sign up to comment on this paper

Sign Up