INFORMATION FLOW THROUGH U-NETS

被引：3

作者：

Lee, Suemin ^{[1
]}

Bajic, Ivan, V ^{[1
]}

机构：

[1] Simon Fraser Univ, Sch Engn Sci, Burnaby, BC, Canada

来源：

2021 IEEE 18TH INTERNATIONAL SYMPOSIUM ON BIOMEDICAL IMAGING (ISBI) | 2021年

基金：

加拿大自然科学与工程研究理事会;

关键词：

U-Net; image segmentation; information flow; mutual information; U-Plot;

D O I：

10.1109/ISBI48211.2021.9433801

中图分类号：

R318 [生物医学工程];

学科分类号：

0831 ;

摘要：

Deep Neural Networks (DNNs) have become ubiquitous in medical image processing and analysis. Among them, U-Nets are very popular in various image segmentation tasks. Yet, little is known about how information flows through these networks and whether they are indeed properly designed for the tasks they are being proposed for. In this paper, we employ information-theoretic tools in order to gain insight into information flow through U-Nets. In particular, we show how mutual information between input/output and an intermediate layer can be a useful tool to understand information flow through various portions of a U-Net, assess its architectural efficiency, and even propose more efficient designs.

引用

页码：812 / 816

页数：5

共 50 条

[31] Improving wildland fire spread prediction using deep U-Nets
Khennou, Fadoua
Akhloufi, Moulay A.
SCIENCE OF REMOTE SENSING, 2023, 8
[32] Multilevel Superpixel Structured Graph U-Nets for Hyperspectral Image Classification
Liu, Qichao
Xiao, Liang
Yang, Jingxiang
Wei, Zhihui
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2022, 60
[33] Automatic Pollen Classification and Segmentation Using U-Nets and Synthetic Data
Boldeanu, Mihai
Gonzalez-Alonso, Monica
Cucu, Horia
Burileanu, Corneliu
Maya-Manzano, Jose Maria
Buters, Jeroen Titus Maria
IEEE ACCESS, 2022, 10 : 73675 - 73684
[34] Binarization of degraded document images with global-local U-Nets
Huang, Xiao
Li, Lin
Liu, Rong
Xu, Chengshen
Ye, Mingdeng
OPTIK, 2020, 203 (203):
[35] Stacked U-Nets for Ground Material Segmentation in Remote Sensing Imagery
Ghosh, Arthita
Ehrlich, Max
Shah, Sohil
Davis, Larry
Chellappa, Rama
PROCEEDINGS 2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION WORKSHOPS (CVPRW), 2018, : 252 - 256
[36] Contrast enhancement of orthopantomograms to improve tooth segmentation using U-Nets
Gonzalez Aseretto, Sebastian
Vazquez Noguera, Jose Luis
2022 XVLIII LATIN AMERICAN COMPUTER CONFERENCE (CLEI 2022), 2022,
[37] Stroke Thrombus Segmentation on SWAN with Multi-Directional U-Nets
Kobold, J.
Vigneron, V.
Maaref, H.
Fourer, D.
Aghasaryan, M.
Alecu, C.
Chausson, N.
L'Hermitte, Y.
Smadja, D.
Lang, E.
Tome, A. M.
2019 NINTH INTERNATIONAL CONFERENCE ON IMAGE PROCESSING THEORY, TOOLS AND APPLICATIONS (IPTA), 2019,
[38] Speech enhancement using U-nets with wide-context units
Grzywalski, Tomasz
Drgas, Szymon
MULTIMEDIA TOOLS AND APPLICATIONS, 2022, 81 (13) : 18617 - 18639
[39] Mesh U-Nets for 3D Cardiac Deformation Modeling
Beetz, Marcel
Acero, Jorge Corral
Banerjee, Abhirup
Eitel, Ingo
Zacur, Ernesto
Lange, Torben
Stiermaier, Thomas
Evertz, Ruben
Backhaus, Soeren J.
Thiele, Holger
Bueno-Orovio, Alfonso
Lamata, Pablo
Schuster, Andreas
Grau, Vicente
STATISTICAL ATLASES AND COMPUTATIONAL MODELS OF THE HEART: REGULAR AND CMRXMOTION CHALLENGE PAPERS, STACOM 2022, 2022, 13593 : 245 - 257
[40] A Multi-Resolution Framework for U-Nets with Applications to Hierarchical VAEs
Falck, Fabian
Williams, Christopher
Danks, Dominic
Deligiannidis, George
Yau, Christopher
Holmes, Chris
Doucet, Arnaud
Willetts, Matthew
ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 35 (NEURIPS 2022), 2022,

← 1 2 3 4 5 →