Balancing Bounding Box and Mask Annotations for Semi-Supervised Instance Segmentation

dc.contributor.authorTolstykh, Daniil
dc.contributor.authorSlutskiy, Dmitriy
dc.contributor.editorSkala, Václav
dc.date.accessioned2025-07-30T09:18:15Z
dc.date.available2025-07-30T09:18:15Z
dc.date.issued2025
dc.description.abstract-translatedInstance segmentation models are crucial for precise object detection but often require expensive pixel-wise mask annotations. This paper studies the impact of combining bounding box and mask annotations in semi-supervised segmentation. We propose a method that leverages from both types of labeled data within a unified training framework. Through experiments on YOLO (convolution-based) and DETR (transformer-based) architectures, we demonstrate that balancing these annotation types significantly enhances performance while reducing labeling costs, particularly in terms of manual annotation time. Additionally, we evaluate few-shot and zero-shot scenarios, further highlighting the flexibility and efficiency of our method for budget-constrained segmentation tasks.en
dc.description.sponsorshipThe authors express their gratitude to Irene De Teresa Trueba and the anonymous referees for their valuable comments, which significantly enhanced the readability of this text. Additionally, the authors extend their thanks to Fabrice Boudaud and the CSAI Lab of ENGIE CRIGEN for their support in this work.
dc.format12 s.cs
dc.format.mimetypeapplication/pdf
dc.identifier.doihttp://www.doi.org/10.24132/CSRN.2025-9
dc.identifier.issn2464-4617 (Print)
dc.identifier.issn2464-4625 (online)
dc.identifier.urihttp://hdl.handle.net/11025/62215
dc.language.isoenen
dc.publisherVaclav Skala - UNION Agencyen
dc.rights© Vaclav Skala - UNION Agencyen
dc.rights.accessopenAccessen
dc.subjectsegmentace instancícs
dc.subjectpolo-supervizované učenícs
dc.subjectoptimalizace nákladů na označovánícs
dc.subjectYOLOv5cs
dc.subjectDETRcs
dc.subject.translatedinstance segmentationen
dc.subject.translatedsemi-supervised learningen
dc.subject.translatedlabeling cost optimizationen
dc.subject.translatedYOLOv5en
dc.subject.translatedDETRen
dc.titleBalancing Bounding Box and Mask Annotations for Semi-Supervised Instance Segmentationen
dc.typekonferenční příspěvekcs
dc.typeconferenceObjecten
dc.type.statusPeer revieweden
dc.type.versionpublishedVersionen
local.files.count1*
local.files.size8313401*
local.has.filesyes*

Files

Original bundle
Showing 1 - 1 out of 1 results
No Thumbnail Available
Name:
B17.pdf
Size:
7.93 MB
Format:
Adobe Portable Document Format
License bundle
Showing 1 - 1 out of 1 results
No Thumbnail Available
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description: