Using different loss functions with YOLACT++ for real-time instance segmentation
dc.contributor.author | Köleş, Selin | |
dc.contributor.author | Karakaş, Selami | |
dc.contributor.author | Ndigande, Alain Patrick | |
dc.contributor.author | Özer, Sedat | |
dc.date.accessioned | 2024-01-23T10:09:32Z | |
dc.date.available | 2024-01-23T10:09:32Z | |
dc.date.issued | 2023 | |
dc.identifier.isbn | 979-835030396-4 | |
dc.identifier.uri | http://hdl.handle.net/10679/9062 | |
dc.identifier.uri | https://ieeexplore.ieee.org/document/10197832 | |
dc.description.abstract | In this paper, we study and analyze the performance of various loss functions on a recently proposed real-time instance segmentation algorithm, YOLACT++. In particular, we study the loss functions, including Huber Loss, Binary Cross Entropy (BCE), Mean Square Error (MSE), Log-Cosh-Dice Loss, and their various combinations within the YOLACT++ architecture. We demonstrate that we can use different loss functions from the default loss function (BCE) of YOLACT++ for improved real-time segmentation results. In our experiments, we show that a certain combination of two loss functions improves the segmentation performance of YOLACT++ in terms of the mean Average Precision (mAP) metric on Cigarettes dataset, when compared to its original loss function. | en_US |
dc.language.iso | eng | en_US |
dc.publisher | IEEE | en_US |
dc.relation.ispartof | 2023 46th International Conference on Telecommunications and Signal Processing (TSP) | |
dc.rights | restrictedAccess | |
dc.title | Using different loss functions with YOLACT++ for real-time instance segmentation | en_US |
dc.type | Conference paper | en_US |
dc.publicationstatus | Published | en_US |
dc.contributor.department | Özyeğin University | |
dc.contributor.authorID | (ORCID 0000-0002-2069-3807 & YÖK ID 386309) Özer, Sedat | |
dc.contributor.ozuauthor | Özer, Sedat | |
dc.identifier.startpage | 264 | en_US |
dc.identifier.endpage | 267 | en_US |
dc.identifier.doi | 10.1109/TSP59544.2023.10197832 | en_US |
dc.subject.keywords | Instance segmentation | en_US |
dc.subject.keywords | Loss function | en_US |
dc.subject.keywords | Real time segmentation | en_US |
dc.subject.keywords | YOLACT++ | en_US |
dc.identifier.scopus | SCOPUS:2-s2.0-85168650663 | |
dc.contributor.ozugradstudent | Köleş, Selin | |
dc.contributor.ozugradstudent | Karakaş, Selami | |
dc.contributor.ozugradstudent | Ndigande, Alain Patrick | |
dc.relation.publicationcategory | Conference Paper - International - Institutional Academic Staff, Graduate Student and Undergraduate Student |
Files in this item
Files | Size | Format | View |
---|---|---|---|
There are no files associated with this item. |
This item appears in the following Collection(s)
Share this page