EUVIMEDEUVIMED entdecken
Crossref

Unsupervised Salient Object Detection on Light Field With High-Quality Synthetic Labels

Yanfeng Zheng, Zhong Luo, Ying Cao, Xiaosong Yang, Weiwei Xu, Zheng Lin, Nan Yin, Pengjie Wang

IEEE Transactions on Circuits and Systems for Video Technology · 2025-05 · 35 (5) · 4608-4618

DOI: 10.1109/tcsvt.2024.3514754 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Circuits and Systems for Video Technology, 2025-05. Verfasst von Yanfeng Zheng und 7 weiteren Autor:innen. Die Arbeit wurde bisher 12-mal zitiert (Crossref) und nennt 58 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Most current Light Field Salient Object Detection (LFSOD) methods require full supervision with labor-intensive pixel-level annotations. Unsupervised Light Field Salient Object Detection (ULFSOD) has gained attention due to this limitation. However, existing methods use traditional handcrafted techniques to generate noisy pseudo-labels, which degrades the performance of models trained on them. To mitigate this issue, we present a novel learning-based approach to synthesize labels for ULFSOD. We introduce a prominent focal stack identification module that utilizes light field information (focal stack, depth map, and RGB color image) to generate high-quality pixel-level pseudo-labels, aiding network training. Additionally, we propose a novel model architecture for LFSOD, combining a multi-scale spatial attention module for focal stack information with a cross fusion module for RGB and focal stack integration. Through extensive experiments, we demonstrate that our pseudo-label generation method significantly outperforms existing methods in label quality. Our proposed model, trained with our labels, shows significant improvement on ULFSOD, achieving new state-of-the-art scores across public benchmarks.

Zitieren

APA

Zheng, Y., Luo, Z., Cao, Y., Yang, X., Xu, W., Lin, Z., Yin, N., & Wang, P. (2025). Unsupervised Salient Object Detection on Light Field With High-Quality Synthetic Labels. IEEE Transactions on Circuits and Systems for Video Technology, 35(5), 4608-4618. https://doi.org/10.1109/tcsvt.2024.3514754

Vancouver

Zheng Y, Luo Z, Cao Y, Yang X, Xu W, Lin Z, et al. Unsupervised Salient Object Detection on Light Field With High-Quality Synthetic Labels. IEEE Transactions on Circuits and Systems for Video Technology. 2025;35(5):4608-4618. doi:10.1109/tcsvt.2024.3514754

BibTeX
@article{Zheng2025,
  author = {Zheng, Yanfeng and Luo, Zhong and Cao, Ying and Yang, Xiaosong and Xu, Weiwei and Lin, Zheng and Yin, Nan and Wang, Pengjie},
  title = {Unsupervised Salient Object Detection on Light Field With High-Quality Synthetic Labels},
  journal = {IEEE Transactions on Circuits and Systems for Video Technology},
  year = {2025},
  volume = {35},
  number = {5},
  pages = {4608--4618},
  doi = {10.1109/tcsvt.2024.3514754}
}
Kostenlos starten →