Unsupervised Salient Object Detection on Light Field With High-Quality Synthetic Labels
Zheng, Yanfeng Luo, Zhong Cao, Ying Yang, Xiaosong Xu, Weiwei Lin, Zheng Yin, Nan Wang, Pengjie
Zheng, Yanfeng
Luo, Zhong
Cao, Ying
Yang, Xiaosong
Xu, Weiwei
Lin, Zheng
Yin, Nan
Wang, Pengjie
Supervisor
Department
Machine Learning
Embargo End Date
Type
Journal article
Date
2025
License
Language
English
Collections
Research Projects
Organizational Units
Journal Issue
Abstract
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.
Citation
Y. Zheng et al., "Unsupervised Salient Object Detection on Light Field With High-Quality Synthetic Labels," in IEEE Transactions on Circuits and Systems for Video Technology, vol. 35, no. 5, pp. 4608-4618, May 2025, doi: 10.1109/TCSVT.2024.3514754
Source
IEEE Transactions on Circuits and Systems for Video Technology
Conference
Keywords
Light field, salient object detection, unsupervised model
Subjects
Source
Publisher
IEEE