Benchmarking the robustness of instance segmentation models

Simple item page
buir.contributor.authorAltındiş, Said Fahri
buir.contributor.authorDündar, Ayşegül
buir.contributor.orcidAltındiş, Said Fahri|0009-0005-5527-4986
buir.contributor.orcidDündar, Ayşegül|0000-0003-2014-6325
dc.citation.epage15en_US
dc.citation.spage1
dc.contributor.authorDalva, Y.
dc.contributor.authorPehlivan, H.
dc.contributor.authorAltındiş, Said Fahri
dc.contributor.authorDündar, Ayşegül
dc.date.accessioned2024-03-19T10:11:58Z
dc.date.available2024-03-19T10:11:58Z
dc.date.issued2023-08-29
dc.departmentDepartment of Computer Engineering
dc.description.abstractThis article presents a comprehensive evaluation of instance segmentation models with respect to real-world image corruptions as well as out-of-domain image collections, e.g., images captured by a different set-up than the training dataset. The out-of-domain image evaluation shows the generalization capability of models, an essential aspect of real-world applica tions, and an extensively studied topic of domain adaptation. These presented robustness and generalization evaluations are important when designing instance segmentation models for real-world applications and picking an off-the-shelf pretrained model to directly use for the task at hand. Specifically, this benchmark study includes state-of-the-art network architectures, network backbones, normalization layers, models trained starting from scratch versus pretrained networks, and the effect of multitask training on robustness and generalization. Through this study, we gain several insights. For example, we find that group normalization (GN) enhances the robustness of networks across corruptions where the image contents stay the same but corruptions are added on top. On the other hand, batch normalization (BN) improves the generalization of the models across different datasets where statistics of image features change. We also find that single-stage detectors do not generalize well to larger image resolutions than their training size. On the other hand, multistage detectors can easily be used on images of different sizes. We hope that our comprehensive study will motivate the development of more robust and reliable instance segmentation models.
dc.identifier.doi10.1109/TNNLS.2023.3310985
dc.identifier.eissn2162-2388
dc.identifier.issn2162-237X
dc.identifier.urihttps://hdl.handle.net/11693/114966
dc.language.isoEnglish
dc.publisherInstitute of Electrical and Electronics Engineers
dc.relation.isversionofhttps://dx.doi.org/10.1109/TNNLS.2023.3310985
dc.source.titleIEEE Transactions on Neural Networks and Learning Systems
dc.subjectDeep networks
dc.subjectDomain adaptation
dc.subjectImage corruptions
dc.subjectInstance segmentation
dc.subjectRobustness
dc.titleBenchmarking the robustness of instance segmentation models
dc.typeArticle

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Benchmarking_the_robustness_of_instance_segmentation_models.pdf
Size:
6.43 MB
Format:
Adobe Portable Document Format
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
2.01 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

Department of Computer Engineering