Spatial Mask-Based Adaptive Robust Training for Video Object Segmentation With Noisy Labels
IEEE Transactions on Circuits and Systems for Video Technology · 2025-05 · 35 (5) · 4976-4990
DOI: 10.1109/tcsvt.2025.3525629 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Circuits and Systems for Video Technology, 2025-05. Verfasst von Enki Cho und 2 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 64 Literaturangaben. Gefördert von: Institute of Information and Communications Technology Planning and Evaluation (IITP) through Korean Government, National Research Foundation of Korea (NRF) through Korean Government.
Abstract (Originalsprache)
Recent advances in video object segmentation (VOS) highlight its potential across various applications. Semi-supervised VOS aims to segment target objects in video frames based on annotations from the initial frame. Collecting a large-scale video segmentation dataset is challenging, which could induce noisy labels. However, it has been overlooked and most of the research efforts have been devoted to training VOS models by assuming the training dataset is clean. In this study, we first explore the effect of VOS models under noisy labels in the training dataset. To investigate the effect of noisy labels, we simulate the noisy annotations on DAVIS 2017 and YouTubeVOS datasets. Experiments show that the traditional training strategy is vulnerable to noisy annotations. To address this issue, we propose a novel noise-robust training method, named SMART (Spatial Mask-based Adaptive Robust Training), which is designed to train models effectively in the presence of noisy annotations. The proposed method employs two key strategies. Firstly, the model focuses on the common spatial areas from clean knowledge-based predictions and annotations. Secondly, the model is trained with adaptive balancing losses based on their reliability. Comparative experiments have demonstrated the effectiveness of our approach by outperforming other noise handling methods over various noise degrees.
Förderung
- Institute of Information and Communications Technology Planning and Evaluation (IITP) through Korean Government
- National Research Foundation of Korea (NRF) through Korean Government
Zitieren
APA
Cho, E., Kim, J. U., & Kim, S. T. (2025). Spatial Mask-Based Adaptive Robust Training for Video Object Segmentation With Noisy Labels. IEEE Transactions on Circuits and Systems for Video Technology, 35(5), 4976-4990. https://doi.org/10.1109/tcsvt.2025.3525629
Vancouver
Cho E, Kim JU, Kim ST. Spatial Mask-Based Adaptive Robust Training for Video Object Segmentation With Noisy Labels. IEEE Transactions on Circuits and Systems for Video Technology. 2025;35(5):4976-4990. doi:10.1109/tcsvt.2025.3525629
BibTeX
@article{Cho2025,
author = {Cho, Enki and Kim, Jung Uk and Kim, Seong Tae},
title = {Spatial Mask-Based Adaptive Robust Training for Video Object Segmentation With Noisy Labels},
journal = {IEEE Transactions on Circuits and Systems for Video Technology},
year = {2025},
volume = {35},
number = {5},
pages = {4976--4990},
doi = {10.1109/tcsvt.2025.3525629}
}
EUVIMED