Anomaly Discovery in Semantic Segmentation via Distillation Comparison Networks

12/18/2021
by   Huan Zhou, et al.
0

This paper aims to address the problem of anomaly discovery in semantic segmentation. Our key observation is that semantic classification plays a critical role in existing approaches, while the incorrectly classified pixels are easily regarded as anomalies. Such a phenomenon frequently appears and is rarely discussed, which significantly reduces the performance of anomaly discovery. To this end, we propose a novel Distillation Comparison Network (DiCNet). It comprises of a teacher branch which is a semantic segmentation network that removed the semantic classification head, and a student branch that is distilled from the teacher branch through a distribution distillation. We show that the distillation guarantees the semantic features of the two branches hold consistency in the known classes, while reflect inconsistency in the unknown class. Therefore, we leverage the semantic feature discrepancy between the two branches to discover the anomalies. DiCNet abandons the semantic classification head in the inference process, and hence significantly alleviates the issue caused by incorrect semantic classification. Extensive experimental results on StreetHazards dataset and BDD-Anomaly dataset are conducted to verify the superior performance of DiCNet. In particular, DiCNet obtains a 6.3 StreetHazards dataset, achieves a 4.2 improvement in FPR95 on BDD-Anomaly dataset. Codes are available at https://github.com/zhouhuan-hust/DiCNet.

READ FULL TEXT

page 1

page 6

page 7

research
05/07/2022

Distilling Inter-Class Distance for Semantic Segmentation

Knowledge distillation is widely adopted in semantic segmentation to red...
research
04/14/2022

Cross-Image Relational Knowledge Distillation for Semantic Segmentation

Current Knowledge Distillation (KD) methods for semantic segmentation of...
research
08/08/2023

AICSD: Adaptive Inter-Class Similarity Distillation for Semantic Segmentation

In recent years, deep neural networks have achieved remarkable accuracy ...
research
06/01/2021

Detecting Anomalies in Semantic Segmentation with Prototypes

Traditional semantic segmentation methods can recognize at test time onl...
research
01/30/2023

FractalAD: A simple industrial anomaly segmentation method using fractal anomaly generation and backbone knowledge distillation

Although industrial anomaly detection (AD) technology has made significa...
research
10/19/2022

Asymmetric Distillation Post-Segmentation Method for Image Anomaly Detection

Knowledge distillation-based anomaly detection methods generate same out...
research
12/17/2021

Weakly Supervised Semantic Segmentation via Alternative Self-Dual Teaching

Current weakly supervised semantic segmentation (WSSS) frameworks usuall...

Please sign up or login with your details

Forgot password? Click here to reset