Residual pattern learning for pixel-wise out-of-distribution detection in semantic segmentation

Semantic segmentation models classify pixels into a set of known ("in-distribution") visual classes. When deployed in an open world, the reliability of these models depends on their ability to not only classify in-distribution pixels but also to detect out-of-distribution (OoD) pixels. His...

Full description

Saved in:
Bibliographic Details
Main Authors: LIU, Y, DING, Choubo, TIAN, Yu, PANG, Guansong, BELAGIANNIS, Vasileios, REID, Ian, CARNEIRO, Gustavo
Format: text
Language:English
Published: Institutional Knowledge at Singapore Management University 2023
Subjects:
Online Access:https://ink.library.smu.edu.sg/sis_research/8413
https://ink.library.smu.edu.sg/context/sis_research/article/9416/viewcontent/Liu_Residual_Pattern_Learning_for_Pixel_Wise_Out_of_Distribution_Detection_in_Semantic_Segmentation_ICCV_2023_paper.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Singapore Management University
Language: English
Description
Summary:Semantic segmentation models classify pixels into a set of known ("in-distribution") visual classes. When deployed in an open world, the reliability of these models depends on their ability to not only classify in-distribution pixels but also to detect out-of-distribution (OoD) pixels. Historically, the poor OoD detection performance of these models has motivated the design of methods based on model re-training using synthetic training images that include OoD visual objects. Although successful, these re-trained methods have two issues: 1) their in-distribution segmentation accuracy may drop during re-training, and 2) their OoD detection accuracy does not generalise well to new contexts (e.g., country surroundings) outside the training set (e.g., city surroundings). In this paper, we mitigate these issues with: (i) a new residual pattern learning (RPL) module that assists the segmentation model to detect OoD pixels with minimal deterioration to the inlier segmentation performance; and (ii) a novel context-robust contrastive learning (CoroCL) that enforces RPL to robustly detect OoD pixels in various contexts. Our approach improves by around 10% FPR and 7% AuPRC the previous state-of-the-art in Fishyscapes, Segment-Me-If-You-Can, and RoadAnomaly datasets. Code will be available.