Multi-level OOD detection in image segmentation networks for safety in automotive applications

This research paper aims to study the efficacy of utilizing multi-layer outputs from semantic segmentation models for handling various types of out-of-distribution (OOD) samples. Specifically, our focus lies on the comparison between early and late layers to determine which layer proves more e...

Full description

Saved in:
Bibliographic Details
Main Author: Poh, Eugene Yang Quan
Other Authors: Arvind Easwaran
Format: Final Year Project
Language:English
Published: Nanyang Technological University 2024
Subjects:
Online Access:https://hdl.handle.net/10356/175033
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-175033
record_format dspace
spelling sg-ntu-dr.10356-1750332024-04-19T15:45:05Z Multi-level OOD detection in image segmentation networks for safety in automotive applications Poh, Eugene Yang Quan Arvind Easwaran School of Computer Science and Engineering arvinde@ntu.edu.sg Computer and Information Science Semantic segmentation OOD detection This research paper aims to study the efficacy of utilizing multi-layer outputs from semantic segmentation models for handling various types of out-of-distribution (OOD) samples. Specifically, our focus lies on the comparison between early and late layers to determine which layer proves more effective in addressing the different kinds of OOD categories. In pursuit of this objective, we analyze the output generated by different layers of the semantic segmentation model and assess their influence on the OOD detection performance when confronted with diverse OOD scenarios. Our study encompasses an examination of two primary OOD types, namely samples with covariate shift (e.g., white noise) and semantically OOD samples. The outcomes of this research aim to contribute valuable insights into the optimal utilization of multi-layer information in semantic segmentation models for improved OOD handling. By discerning the strengths and weaknesses of early and late layers in the context of distinct OOD challenges, we seek to provide a nuanced understanding of the interplay between model architecture and OOD sample characteristics. Ultimately, we hope that our findings may pave the way for enhanced model robustness and generalization capabilities in real-world scenarios. Bachelor's degree 2024-04-18T23:31:57Z 2024-04-18T23:31:57Z 2024 Final Year Project (FYP) Poh, E. Y. Q. (2024). Multi-level OOD detection in image segmentation networks for safety in automotive applications. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/175033 https://hdl.handle.net/10356/175033 en application/pdf Nanyang Technological University
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Computer and Information Science
Semantic segmentation
OOD detection
spellingShingle Computer and Information Science
Semantic segmentation
OOD detection
Poh, Eugene Yang Quan
Multi-level OOD detection in image segmentation networks for safety in automotive applications
description This research paper aims to study the efficacy of utilizing multi-layer outputs from semantic segmentation models for handling various types of out-of-distribution (OOD) samples. Specifically, our focus lies on the comparison between early and late layers to determine which layer proves more effective in addressing the different kinds of OOD categories. In pursuit of this objective, we analyze the output generated by different layers of the semantic segmentation model and assess their influence on the OOD detection performance when confronted with diverse OOD scenarios. Our study encompasses an examination of two primary OOD types, namely samples with covariate shift (e.g., white noise) and semantically OOD samples. The outcomes of this research aim to contribute valuable insights into the optimal utilization of multi-layer information in semantic segmentation models for improved OOD handling. By discerning the strengths and weaknesses of early and late layers in the context of distinct OOD challenges, we seek to provide a nuanced understanding of the interplay between model architecture and OOD sample characteristics. Ultimately, we hope that our findings may pave the way for enhanced model robustness and generalization capabilities in real-world scenarios.
author2 Arvind Easwaran
author_facet Arvind Easwaran
Poh, Eugene Yang Quan
format Final Year Project
author Poh, Eugene Yang Quan
author_sort Poh, Eugene Yang Quan
title Multi-level OOD detection in image segmentation networks for safety in automotive applications
title_short Multi-level OOD detection in image segmentation networks for safety in automotive applications
title_full Multi-level OOD detection in image segmentation networks for safety in automotive applications
title_fullStr Multi-level OOD detection in image segmentation networks for safety in automotive applications
title_full_unstemmed Multi-level OOD detection in image segmentation networks for safety in automotive applications
title_sort multi-level ood detection in image segmentation networks for safety in automotive applications
publisher Nanyang Technological University
publishDate 2024
url https://hdl.handle.net/10356/175033
_version_ 1800916377585844224