DC Field | Value | Language |
---|---|---|
dc.contributor.author | KIM, HOYOUNG | - |
dc.contributor.author | MINHYEON, OH | - |
dc.contributor.author | HWANG, SEHYUN | - |
dc.contributor.author | KWAK, SU HA | - |
dc.contributor.author | OK, JUNGSEUL | - |
dc.date.accessioned | 2024-01-23T00:25:17Z | - |
dc.date.available | 2024-01-23T00:25:17Z | - |
dc.date.created | 2023-12-19 | - |
dc.date.issued | 2023-10-04 | - |
dc.identifier.uri | https://oasis.postech.ac.kr/handle/2014.oak/119818 | - |
dc.description.abstract | Learning semantic segmentation requires pixel-wise annotations, which can be time-consuming and expensive. To reduce the annotation cost, we propose a superpixel-based active learning (AL) framework, which collects a dominant label per superpixel instead. To be specific, it consists of adaptive superpixel and sieving mechanisms, fully dedicated to AL. At each round of AL, we adaptively merge neighboring pixels of similar learned features into superpixels. We then query a selected subset of these superpixels using an acquisition function assuming no uniform superpixel size. This approach is more efficient than existing methods, which rely only on innate features such as RGB color and assume uniform superpixel sizes. Obtaining a dominant label per superpixel drastically reduces annotators' burden as it requires fewer clicks. However, it inevitably introduces noisy annotations due to mismatches between superpixel and ground truth segmentation. To address this issue, we further devise a sieving mechanism that identifies and excludes potentially noisy annotations from learning. Our experiments on both Cityscapes and PASCAL VOC datasets demonstrate the efficacy of adaptive superpixel and sieving mechanisms. | - |
dc.language | English | - |
dc.publisher | Computer Vision Foundation | - |
dc.relation.isPartOf | International Conference on Computer Vision 2023 | - |
dc.title | Adaptive Superpixel for Active Learning in Semantic Segmentation | - |
dc.type | Conference | - |
dc.type.rims | CONF | - |
dc.identifier.bibliographicCitation | International Conference on Computer Vision 2023 | - |
dc.citation.conferenceDate | 2023-10-02 | - |
dc.citation.conferencePlace | FR | - |
dc.citation.title | International Conference on Computer Vision 2023 | - |
dc.contributor.affiliatedAuthor | KIM, HOYOUNG | - |
dc.contributor.affiliatedAuthor | MINHYEON, OH | - |
dc.contributor.affiliatedAuthor | HWANG, SEHYUN | - |
dc.contributor.affiliatedAuthor | KWAK, SU HA | - |
dc.contributor.affiliatedAuthor | OK, JUNGSEUL | - |
dc.description.journalClass | 1 | - |
dc.description.journalClass | 1 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.
library@postech.ac.kr Tel: 054-279-2548
Copyrights © by 2017 Pohang University of Science ad Technology All right reserved.