Matches in SemOpenAlex for { <https://semopenalex.org/work/W3197579076> ?p ?o ?g. }
- W3197579076 endingPage "15" @default.
- W3197579076 startingPage "1" @default.
- W3197579076 abstract "This 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 applications, 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." @default.
- W3197579076 created "2021-09-13" @default.
- W3197579076 creator A5003803677 @default.
- W3197579076 creator A5041525280 @default.
- W3197579076 creator A5083169090 @default.
- W3197579076 date "2023-01-01" @default.
- W3197579076 modified "2023-09-27" @default.
- W3197579076 title "Benchmarking the Robustness of Instance Segmentation Models" @default.
- W3197579076 cites W1536680647 @default.
- W3197579076 cites W1565327149 @default.
- W3197579076 cites W1722318740 @default.
- W3197579076 cites W1836465849 @default.
- W3197579076 cites W1861492603 @default.
- W3197579076 cites W2102605133 @default.
- W3197579076 cites W2109255472 @default.
- W3197579076 cites W2159291411 @default.
- W3197579076 cites W2180290696 @default.
- W3197579076 cites W2194775991 @default.
- W3197579076 cites W2334805829 @default.
- W3197579076 cites W2340897893 @default.
- W3197579076 cites W2342045095 @default.
- W3197579076 cites W2502312327 @default.
- W3197579076 cites W2549139847 @default.
- W3197579076 cites W2562192638 @default.
- W3197579076 cites W2565639579 @default.
- W3197579076 cites W2593768305 @default.
- W3197579076 cites W2601564443 @default.
- W3197579076 cites W2605102758 @default.
- W3197579076 cites W2613718673 @default.
- W3197579076 cites W2795783309 @default.
- W3197579076 cites W2888339491 @default.
- W3197579076 cites W2901394229 @default.
- W3197579076 cites W2905744087 @default.
- W3197579076 cites W2950468330 @default.
- W3197579076 cites W2952610664 @default.
- W3197579076 cites W2961540362 @default.
- W3197579076 cites W2962793481 @default.
- W3197579076 cites W2963037989 @default.
- W3197579076 cites W2963060032 @default.
- W3197579076 cites W2963107255 @default.
- W3197579076 cites W2963150697 @default.
- W3197579076 cites W2963201472 @default.
- W3197579076 cites W2963207607 @default.
- W3197579076 cites W2963542991 @default.
- W3197579076 cites W2963709863 @default.
- W3197579076 cites W2963739978 @default.
- W3197579076 cites W2963826681 @default.
- W3197579076 cites W2963840672 @default.
- W3197579076 cites W2963850211 @default.
- W3197579076 cites W2963857746 @default.
- W3197579076 cites W2963998105 @default.
- W3197579076 cites W2964097310 @default.
- W3197579076 cites W2964153729 @default.
- W3197579076 cites W2964241181 @default.
- W3197579076 cites W2964288524 @default.
- W3197579076 cites W2964294232 @default.
- W3197579076 cites W2964309882 @default.
- W3197579076 cites W2964341837 @default.
- W3197579076 cites W2966926453 @default.
- W3197579076 cites W2989676862 @default.
- W3197579076 cites W3002450089 @default.
- W3197579076 cites W3035087309 @default.
- W3197579076 cites W3035564946 @default.
- W3197579076 cites W3035755044 @default.
- W3197579076 cites W3094984771 @default.
- W3197579076 cites W3103259849 @default.
- W3197579076 cites W3105724776 @default.
- W3197579076 cites W3106250896 @default.
- W3197579076 cites W3106297436 @default.
- W3197579076 cites W3108281323 @default.
- W3197579076 cites W3126058723 @default.
- W3197579076 cites W3137905681 @default.
- W3197579076 cites W3176659256 @default.
- W3197579076 cites W3197074992 @default.
- W3197579076 doi "https://doi.org/10.1109/tnnls.2023.3310985" @default.
- W3197579076 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37721888" @default.
- W3197579076 hasPublicationYear "2023" @default.
- W3197579076 type Work @default.
- W3197579076 sameAs 3197579076 @default.
- W3197579076 citedByCount "1" @default.
- W3197579076 countsByYear W31975790762021 @default.
- W3197579076 crossrefType "journal-article" @default.
- W3197579076 hasAuthorship W3197579076A5003803677 @default.
- W3197579076 hasAuthorship W3197579076A5041525280 @default.
- W3197579076 hasAuthorship W3197579076A5083169090 @default.
- W3197579076 hasConcept C104317684 @default.
- W3197579076 hasConcept C119857082 @default.
- W3197579076 hasConcept C134306372 @default.
- W3197579076 hasConcept C136886441 @default.
- W3197579076 hasConcept C144024400 @default.
- W3197579076 hasConcept C144133560 @default.
- W3197579076 hasConcept C153180895 @default.
- W3197579076 hasConcept C154945302 @default.
- W3197579076 hasConcept C162853370 @default.
- W3197579076 hasConcept C177148314 @default.
- W3197579076 hasConcept C185592680 @default.
- W3197579076 hasConcept C19165224 @default.
- W3197579076 hasConcept C33923547 @default.