Weakly-supervised salient object detection (WSOD) aims to develop saliency models using image-level annotations.

Dec 28, 2023 · Weakly-supervised salient object detection (WSOD) aims to develop saliency models using image-level annotations.

Dec 28, 2023 · We propose a self-calibrated training strategy by explicitly establishing a mutual calibration loop between pseudo labels and network predictions.

Dec 31, 2023 · Comprehensive experiments demonstrate that our method outperforms all the existing WSOD methods by adopting the self-calibrated strategy only.

Source code for the Paper: "To be Critical: Self-Calibrated Weakly Supervised Learning for Salient Object Detection. " Jian Wang, Miao Zhang, Yongri Piao ...

Sep 4, 2021 · Abstract—Weakly-supervised salient object detection (WSOD1) aims to develop saliency models using image-level annotations.

This work proposes a self-calibrated training strategy by explicitly establishing a mutual calibration loop between pseudo labels and network predictions, ...

Apr 25, 2024 · To be Critical: Self-Calibrated Weakly Supervised Learning for Salient Object Detection. CoRR abs/2109.01770 (2021). [+][–]. Coauthor network.

Co-authors ; To Be Critical: Self-calibrated Weakly Supervised Learning for Salient Object Detection. J Wang, T Liu, M Zhang, Y Piao. Chinese Conference on ...

To be Critical: Self-Calibrated Weakly Supervised Learning for Salient Object Detection · no code implementations • 4 Sep 2021 • Yongri Piao, Jian Wang, Miao ...