ConformalSAM: Unlocking the Potential of Foundational Segmentation Models in Semi-Supervised Semantic Segmentation with Conformal Prediction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Danhui, Liu, Ziquan, Yang, Chuxi, Wang, Dan, Yan, Yan, Xu, Yi, Ji, Xiangyang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913951369920512
author Chen, Danhui
Liu, Ziquan
Yang, Chuxi
Wang, Dan
Yan, Yan
Xu, Yi
Ji, Xiangyang
author_facet Chen, Danhui
Liu, Ziquan
Yang, Chuxi
Wang, Dan
Yan, Yan
Xu, Yi
Ji, Xiangyang
contents Pixel-level vision tasks, such as semantic segmentation, require extensive and high-quality annotated data, which is costly to obtain. Semi-supervised semantic segmentation (SSSS) has emerged as a solution to alleviate the labeling burden by leveraging both labeled and unlabeled data through self-training techniques. Meanwhile, the advent of foundational segmentation models pre-trained on massive data, has shown the potential to generalize across domains effectively. This work explores whether a foundational segmentation model can address label scarcity in the pixel-level vision task as an annotator for unlabeled images. Specifically, we investigate the efficacy of using SEEM, a Segment Anything Model (SAM) variant fine-tuned for textual input, to generate predictive masks for unlabeled data. To address the shortcomings of using SEEM-generated masks as supervision, we propose ConformalSAM, a novel SSSS framework which first calibrates the foundation model using the target domain's labeled data and then filters out unreliable pixel labels of unlabeled data so that only high-confidence labels are used as supervision. By leveraging conformal prediction (CP) to adapt foundation models to target data through uncertainty calibration, ConformalSAM exploits the strong capability of the foundational segmentation model reliably which benefits the early-stage learning, while a subsequent self-reliance training strategy mitigates overfitting to SEEM-generated masks in the later training stage. Our experiment demonstrates that, on three standard benchmarks of SSSS, ConformalSAM achieves superior performance compared to recent SSSS methods and helps boost the performance of those methods as a plug-in.
format Preprint
id arxiv_https___arxiv_org_abs_2507_15803
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ConformalSAM: Unlocking the Potential of Foundational Segmentation Models in Semi-Supervised Semantic Segmentation with Conformal Prediction
Chen, Danhui
Liu, Ziquan
Yang, Chuxi
Wang, Dan
Yan, Yan
Xu, Yi
Ji, Xiangyang
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Pixel-level vision tasks, such as semantic segmentation, require extensive and high-quality annotated data, which is costly to obtain. Semi-supervised semantic segmentation (SSSS) has emerged as a solution to alleviate the labeling burden by leveraging both labeled and unlabeled data through self-training techniques. Meanwhile, the advent of foundational segmentation models pre-trained on massive data, has shown the potential to generalize across domains effectively. This work explores whether a foundational segmentation model can address label scarcity in the pixel-level vision task as an annotator for unlabeled images. Specifically, we investigate the efficacy of using SEEM, a Segment Anything Model (SAM) variant fine-tuned for textual input, to generate predictive masks for unlabeled data. To address the shortcomings of using SEEM-generated masks as supervision, we propose ConformalSAM, a novel SSSS framework which first calibrates the foundation model using the target domain's labeled data and then filters out unreliable pixel labels of unlabeled data so that only high-confidence labels are used as supervision. By leveraging conformal prediction (CP) to adapt foundation models to target data through uncertainty calibration, ConformalSAM exploits the strong capability of the foundational segmentation model reliably which benefits the early-stage learning, while a subsequent self-reliance training strategy mitigates overfitting to SEEM-generated masks in the later training stage. Our experiment demonstrates that, on three standard benchmarks of SSSS, ConformalSAM achieves superior performance compared to recent SSSS methods and helps boost the performance of those methods as a plug-in.
title ConformalSAM: Unlocking the Potential of Foundational Segmentation Models in Semi-Supervised Semantic Segmentation with Conformal Prediction
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2507.15803