TY - GEN
T1 - Cartoon Explanations of Image Classifiers
AU - Kolek, Stefan
AU - Nguyen, Duc Anh
AU - Levie, Ron
AU - Bruna, Joan
AU - Kutyniok, Gitta
N1 - Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
PY - 2022
Y1 - 2022
N2 - We present CartoonX (Cartoon Explanation), a novel model-agnostic explanation method tailored towards image classifiers and based on the rate-distortion explanation (RDE) framework. Natural images are roughly piece-wise smooth signals—also called cartoon-like images—and tend to be sparse in the wavelet domain. CartoonX is the first explanation method to exploit this by requiring its explanations to be sparse in the wavelet domain, thus extracting the relevant piece-wise smooth part of an image instead of relevant pixel-sparse regions. We demonstrate that CartoonX can reveal novel valuable explanatory information, particularly for misclassifications. Moreover, we show that CartoonX achieves a lower distortion with fewer coefficients than state-of-the-art methods.
AB - We present CartoonX (Cartoon Explanation), a novel model-agnostic explanation method tailored towards image classifiers and based on the rate-distortion explanation (RDE) framework. Natural images are roughly piece-wise smooth signals—also called cartoon-like images—and tend to be sparse in the wavelet domain. CartoonX is the first explanation method to exploit this by requiring its explanations to be sparse in the wavelet domain, thus extracting the relevant piece-wise smooth part of an image instead of relevant pixel-sparse regions. We demonstrate that CartoonX can reveal novel valuable explanatory information, particularly for misclassifications. Moreover, we show that CartoonX achieves a lower distortion with fewer coefficients than state-of-the-art methods.
UR - http://www.scopus.com/inward/record.url?scp=85142759363&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85142759363&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-19775-8_26
DO - 10.1007/978-3-031-19775-8_26
M3 - Conference contribution
AN - SCOPUS:85142759363
SN - 9783031197741
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 443
EP - 458
BT - Computer Vision – ECCV 2022 - 17th European Conference, Proceedings
A2 - Avidan, Shai
A2 - Brostow, Gabriel
A2 - Cissé, Moustapha
A2 - Farinella, Giovanni Maria
A2 - Hassner, Tal
PB - Springer Science and Business Media Deutschland GmbH
T2 - 17th European Conference on Computer Vision, ECCV 2022
Y2 - 23 October 2022 through 27 October 2022
ER -