Heatmap-based Out-of-Distribution Detection

11/15/2022
by   Julia Hornauer, et al.
10

Our work investigates out-of-distribution (OOD) detection as a neural network output explanation problem. We learn a heatmap representation for detecting OOD images while visualizing in- and out-of-distribution image regions at the same time. Given a trained and fixed classifier, we train a decoder neural network to produce heatmaps with zero response for in-distribution samples and high response heatmaps for OOD samples, based on the classifier features and the class prediction. Our main innovation lies in the heatmap definition for an OOD sample, as the normalized difference from the closest in-distribution sample. The heatmap serves as a margin to distinguish between in- and out-of-distribution samples. Our approach generates the heatmaps not only for OOD detection, but also to indicate in- and out-of-distribution regions of the input image. In our evaluations, our approach mostly outperforms the prior work on fixed classifiers, trained on CIFAR-10, CIFAR-100 and Tiny ImageNet. The code is publicly available at: https://github.com/jhornauer/heatmap_ood.

READ FULL TEXT

page 1

page 6

page 8

research
03/25/2023

SIO: Synthetic In-Distribution Data Benefits Out-of-Distribution Detection

Building up reliable Out-of-Distribution (OOD) detectors is challenging,...
research
02/23/2023

A framework for benchmarking class-out-of-distribution detection and its application to ImageNet

When deployed for risk-sensitive tasks, deep neural networks must be abl...
research
10/09/2019

Out-of-distribution Detection in Classifiers via Generation

By design, discriminatively trained neural network classifiers produce r...
research
08/15/2023

DiffGuard: Semantic Mismatch-Guided Out-of-Distribution Detection using Pre-trained Diffusion Models

Given a classifier, the inherent property of semantic Out-of-Distributio...
research
03/29/2022

Proactive Image Manipulation Detection

Image manipulation detection algorithms are often trained to discriminat...
research
10/04/2022

CFL-Net: Image Forgery Localization Using Contrastive Learning

Conventional forgery localizing methods usually rely on different forger...
research
11/25/2022

MAEDAY: MAE for few and zero shot AnomalY-Detection

The goal of Anomaly-Detection (AD) is to identify outliers, or outlying ...

Please sign up or login with your details

Forgot password? Click here to reset