Concurrent Misclassification and Out-of-Distribution Detection for Semantic Segmentation via Energy-Based Normalizing Flow

Denis A Gudovskiy, Tomoyuki Okuno, Yohei Nakata

Keywords: semantic segmentation, out-of-distribution, misclassification, generative models, normalizing flows, open-set, robustness

TL;DR: A normalizing flow model with the energy-based inputs to detect in-distribution misclassifications and out-of-distribution examples for semantic segmentation application

Abstract: Recent semantic segmentation models accurately classify test-time examples that are similar to a training dataset distribution. However, their discriminative closed-set approach is not robust in practical data setups with distributional shifts and out-of-distribution (OOD) classes. As a result, the predicted probabilities can be very imprecise when used as confidence scores at test time. To address this, we propose a generative model for concurrent in-distribution misclassification (IDM) and OOD detection that relies on a normalizing flow framework. The proposed flow-based detector with an energy-based inputs (FlowEneDet) can extend previously deployed segmentation models without their time-consuming retraining. Our FlowEneDet results in a low-complexity architecture with marginal increase in the memory footprint. FlowEneDet achieves promising results on Cityscapes, Cityscapes-C, FishyScapes and SegmentMeIfYouCan benchmarks in IDM/OOD detection when applied to pretrained DeepLabV3+ and SegFormer semantic segmentation models.

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/concurrent-misclassification-and-out-of/code)