Robust and Consistent Online Video Instance Segmentation via Instance Mask Propagation
Proceedings of the AAAI Conference on Artificial Intelligence · 11.04.2025 · 39 (4) · 3483-3490
DOI: 10.1609/aaai.v39i4.32361 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 11.04.2025. Verfasst von Miran Heo und 3 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref).
Abstract (Originalsprache)
Recent advancements in online Video Instance Segmentation (VIS) methods show notable performance improvements across benchmarks. However, the leading methods in the tracking-by-detection paradigm often result in temporally inconsistent predictions at both instance-level and pixel-level that lead to visually unsatisfactory outcomes. To address these challenges, we propose RoCoVIS, a simple yet effective approach that integrates segmentation and tracking to provide consistent online VIS. Our approach is an end-to-end sequential learning where object queries are propagated through mask predictions, improving the accuracy of temporal instance mapping at the pixel level. Additionally, we propose a new label assignment criterion in harmony with our approach. We also examine the limitations and challenges presented by the current standard evaluation protocol (AP) and suggest adopting additional metrics, Tube-Boundary AP and AP_Pool. RoCoVIS demonstrates superior performance on challenging VIS benchmarks with a Swin-L backbone and shows competitive results when employing a ResNet-50 backbone. By employing Tube-Boundary AP and AP_Pool as metrics to measure mask accuracy and consistency, RoCoVIS outperforms its counterpart, GenVIS, on the HQ-YTVIS and VIPSeg.
Zitieren
APA
Heo, M., Oh, S. W., Kim, S. J., & Lee, J. (2025). Robust and Consistent Online Video Instance Segmentation via Instance Mask Propagation. Proceedings of the AAAI Conference on Artificial Intelligence, 39(4), 3483-3490. https://doi.org/10.1609/aaai.v39i4.32361
Vancouver
Heo M, Oh SW, Kim SJ, Lee J. Robust and Consistent Online Video Instance Segmentation via Instance Mask Propagation. Proceedings of the AAAI Conference on Artificial Intelligence. 2025;39(4):3483-3490. doi:10.1609/aaai.v39i4.32361
BibTeX
@article{Heo2025,
author = {Heo, Miran and Oh, Seoung Wug and Kim, Seon Joo and Lee, Joon-Young},
title = {Robust and Consistent Online Video Instance Segmentation via Instance Mask Propagation},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2025},
volume = {39},
number = {4},
pages = {3483--3490},
doi = {10.1609/aaai.v39i4.32361}
}
EUVIMED