Visual Feature Attribution using Wasserstein GANs

被引:83
作者
Baumgartner, Christian F. [1 ]
Koch, Lisa M. [2 ]
Tezcan, Kerem Can [1 ]
Ang, Jia Xi [1 ]
Konukoglu, Ender [1 ]
机构
[1] Swiss Fed Inst Technol, Comp Vis Lab, Zurich, Switzerland
[2] Swiss Fed Inst Technol, Comp Vis & Geometry Grp, Zurich, Switzerland
来源
2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR) | 2018年
关键词
ALZHEIMERS-DISEASE; PERFUSION ABNORMALITIES; GENETIC INFLUENCES; IMAGE-ANALYSIS; DEMENTIA; ATROPHY;
D O I
10.1109/CVPR.2018.00867
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Attributing the pixels of an input image to a certain category is an important and well-studied problem in computer vision, with applications ranging from weakly supervised localisation to understanding hidden effects in the data. In recent years, approaches based on interpreting a previously trained neural network classifier have become the de facto state-of-the-art and are commonly used on medical as well as natural image datasets. In this paper, we discuss a limitation of these approaches which may lead to only a subset of the category specific features being detected. To address this problem we develop a novel feature attribution technique based on Wasserstein Generative Adversarial Networks (WGAN), which does not suffer from this limitation. We show that our proposed method performs substantially better than the state-of-the-art for visual attribution on a synthetic dataset and on real 3D neuroimaging data from patients with mild cognitive impairment (MCI) and Alzheimer's disease (AD). For AD patients the method produces compellingly realistic disease effect maps which are very close to the observed effects.
引用
收藏
页码:8309 / 8319
页数:11
相关论文
共 67 条
[1]  
[Anonymous], 2017, P IEEE C COMP VIS PA
[2]  
[Anonymous], 2016, PROC 4 INT C LEARN R
[3]  
[Anonymous], 2017, P IEEE C COMP VIS PA
[4]  
[Anonymous], 2017, ARXIV170403296
[5]  
[Anonymous], IEEE T MED IMAGING
[6]  
[Anonymous], P INT C MED IM COMP
[7]  
[Anonymous], MED COMPUTER VISION
[8]  
[Anonymous], WORKSHOP INT C LEARN
[9]  
[Anonymous], 2014, STRIVING SIMPLICITY
[10]  
[Anonymous], 2015, COMPUT SCI