Pixel-wise energy-biased abstention learning for anomaly segmentation on complex urban driving scenes
State-of-the-art (SOTA) anomaly segmentation approaches on complex urban driving scenes explore pixel-wise classification uncertainty learned from outlier exposure, or external reconstruction models. However, previous uncertainty approaches that directly associate high uncertainty to anomaly may som...
Saved in:
Main Authors: | , , , , , |
---|---|
Format: | text |
Language: | English |
Published: |
Institutional Knowledge at Singapore Management University
2022
|
Subjects: | |
Online Access: | https://ink.library.smu.edu.sg/sis_research/7058 https://ink.library.smu.edu.sg/context/sis_research/article/8061/viewcontent/2111.12264.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Singapore Management University |
Language: | English |
id |
sg-smu-ink.sis_research-8061 |
---|---|
record_format |
dspace |
spelling |
sg-smu-ink.sis_research-80612023-08-08T05:18:18Z Pixel-wise energy-biased abstention learning for anomaly segmentation on complex urban driving scenes TIAN, Yu LIU, Yuyuan PANG, Guansong LIU, Fengbei CHEN, Yuanhong CARNEIRO, Gustavo State-of-the-art (SOTA) anomaly segmentation approaches on complex urban driving scenes explore pixel-wise classification uncertainty learned from outlier exposure, or external reconstruction models. However, previous uncertainty approaches that directly associate high uncertainty to anomaly may sometimes lead to incorrect anomaly predictions, and external reconstruction models tend to be too inefficient for real-time self-driving embedded systems. In this paper, we propose a new anomaly segmentation method, named pixel-wise energy-biased abstention learning (PEBAL), that explores pixel-wise abstention learning (AL) with a model that learns an adaptive pixel-level anomaly class, and an energy-based model (EBM) that learns inlier pixel distribution. More specifically, PEBAL is based on a non-trivial joint training of EBM and AL, where EBM is trained to output high-energy for anomaly pixels (from outlier exposure) and AL is trained such that these high-energy pixels receive adaptive low penalty for being included to the anomaly class. We extensively evaluate PEBAL against the SOTA and show that it achieves the best performance across four benchmarks. Code is available at https://github.com/tianyu0207/PEBAL. 2022-10-01T07:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/7058 info:doi/10.1007/978-3-031-19842-7_15 https://ink.library.smu.edu.sg/context/sis_research/article/8061/viewcontent/2111.12264.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University Artificial Intelligence and Robotics Graphics and Human Computer Interfaces |
institution |
Singapore Management University |
building |
SMU Libraries |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
SMU Libraries |
collection |
InK@SMU |
language |
English |
topic |
Artificial Intelligence and Robotics Graphics and Human Computer Interfaces |
spellingShingle |
Artificial Intelligence and Robotics Graphics and Human Computer Interfaces TIAN, Yu LIU, Yuyuan PANG, Guansong LIU, Fengbei CHEN, Yuanhong CARNEIRO, Gustavo Pixel-wise energy-biased abstention learning for anomaly segmentation on complex urban driving scenes |
description |
State-of-the-art (SOTA) anomaly segmentation approaches on complex urban driving scenes explore pixel-wise classification uncertainty learned from outlier exposure, or external reconstruction models. However, previous uncertainty approaches that directly associate high uncertainty to anomaly may sometimes lead to incorrect anomaly predictions, and external reconstruction models tend to be too inefficient for real-time self-driving embedded systems. In this paper, we propose a new anomaly segmentation method, named pixel-wise energy-biased abstention learning (PEBAL), that explores pixel-wise abstention learning (AL) with a model that learns an adaptive pixel-level anomaly class, and an energy-based model (EBM) that learns inlier pixel distribution. More specifically, PEBAL is based on a non-trivial joint training of EBM and AL, where EBM is trained to output high-energy for anomaly pixels (from outlier exposure) and AL is trained such that these high-energy pixels receive adaptive low penalty for being included to the anomaly class. We extensively evaluate PEBAL against the SOTA and show that it achieves the best performance across four benchmarks. Code is available at https://github.com/tianyu0207/PEBAL. |
format |
text |
author |
TIAN, Yu LIU, Yuyuan PANG, Guansong LIU, Fengbei CHEN, Yuanhong CARNEIRO, Gustavo |
author_facet |
TIAN, Yu LIU, Yuyuan PANG, Guansong LIU, Fengbei CHEN, Yuanhong CARNEIRO, Gustavo |
author_sort |
TIAN, Yu |
title |
Pixel-wise energy-biased abstention learning for anomaly segmentation on complex urban driving scenes |
title_short |
Pixel-wise energy-biased abstention learning for anomaly segmentation on complex urban driving scenes |
title_full |
Pixel-wise energy-biased abstention learning for anomaly segmentation on complex urban driving scenes |
title_fullStr |
Pixel-wise energy-biased abstention learning for anomaly segmentation on complex urban driving scenes |
title_full_unstemmed |
Pixel-wise energy-biased abstention learning for anomaly segmentation on complex urban driving scenes |
title_sort |
pixel-wise energy-biased abstention learning for anomaly segmentation on complex urban driving scenes |
publisher |
Institutional Knowledge at Singapore Management University |
publishDate |
2022 |
url |
https://ink.library.smu.edu.sg/sis_research/7058 https://ink.library.smu.edu.sg/context/sis_research/article/8061/viewcontent/2111.12264.pdf |
_version_ |
1779156843120033792 |