Towards Explainable Computer Vision Methods via Uncertainty Activation Map

Seungyoun Shin, Wonho Bae, Junhyug Noh, Sungjoon Choi

Research output: Chapter in Book/Report/Conference proceedingConference contribution


This paper focuses on the problem of highlighting the input image regions that result in increasing predictive uncertainty. In particular, we focus on two types of uncertainty, epistemic and aleatoric, and present an uncertainty activation mapping method that can incorporate both types of uncertainty. To this end, we first utilize a mixture-of-experts model combined with class-activation mapping (CAM). The proposed method is extensively evaluated in two different scenarios: multi-label and artificial noise injection scenarios, where we show that our proposed method can effectively capture uncertain regions.

Original languageEnglish
Title of host publicationPattern Recognition - 7th Asian Conference, ACPR 2023, Proceedings
EditorsHuimin Lu, Michael Blumenstein, Sung-Bae Cho, Cheng-Lin Liu, Yasushi Yagi, Tohru Kamiya
PublisherSpringer Science and Business Media Deutschland GmbH
Number of pages14
ISBN (Print)9783031476334
Publication statusPublished - 2023
Event7th Asian Conference on Pattern Recognition, ACPR 2023 - Kitakyushu, Japan
Duration: 2023 Nov 52023 Nov 8

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume14406 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference7th Asian Conference on Pattern Recognition, ACPR 2023

Bibliographical note

Publisher Copyright:
© 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.


  • Explainable machine learning
  • Uncertainty estimation
  • Visual explanation

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science


Dive into the research topics of 'Towards Explainable Computer Vision Methods via Uncertainty Activation Map'. Together they form a unique fingerprint.

Cite this