Pseudo-Label Quality Decoupling and Correction for Semi-Supervised Instance Segmentation

2025-05-16

AI Summary

In semi-supervised instance segmentation (SSIS), pseudo-label quality is unstable—particularly due to coupled evaluation of category and mask fidelity, causing performance fluctuations. To address this, we propose a Quality-Decoupling and Correction (QDC) framework that enhances pseudo-label reliability at three granularities: instance-, category-, and pixel-level. First, we introduce a novel decoupled assessment mechanism that independently quantifies category and mask quality. Second, we design a dynamic category correction module to mitigate inter-class confusion. Third, we incorporate an uncertainty-aware pixel-wise loss reweighting strategy to suppress erroneous predictions. Integrated with dual-threshold instance filtering and semi-supervised consistency training, our method achieves new state-of-the-art results: +11.6 mAP on COCO with only 1% labeled data and +15.5 mAP on Cityscapes with 5% labeled data.

Technology Category

• Machine Learning: Semi-Supervised Learning
• Computer Vision: Segmentation
• Search and Optimization: Learning to Search

Application Category

• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Semi-Supervised Instance Segmentation (SSIS) involves classifying and grouping image pixels into distinct object instances using limited labeled data. This learning paradigm usually faces a significant challenge of unstable performance caused by noisy pseudo-labels of instance categories and pixel masks. We find that the prevalent practice of filtering instance pseudo-labels assessing both class and mask quality with a single score threshold, frequently leads to compromises in the trade-off between the qualities of class and mask labels. In this paper, we introduce a novel Pseudo-Label Quality Decoupling and Correction (PL-DC) framework for SSIS to tackle the above challenges. Firstly, at the instance level, a decoupled dual-threshold filtering mechanism is designed to decouple class and mask quality estimations for instance-level pseudo-labels, thereby independently controlling pixel classifying and grouping qualities. Secondly, at the category level, we introduce a dynamic instance category correction module to dynamically correct the pseudo-labels of instance categories, effectively alleviating category confusion. Lastly, we introduce a pixel-level mask uncertainty-aware mechanism at the pixel level to re-weight the mask loss for different pixels, thereby reducing the impact of noise introduced by pixel-level mask pseudo-labels. Extensive experiments on the COCO and Cityscapes datasets demonstrate that the proposed PL-DC achieves significant performance improvements, setting new state-of-the-art results for SSIS. Notably, our PL-DC shows substantial gains even with minimal labeled data, achieving an improvement of +11.6 mAP with just 1% COCO labeled data and +15.5 mAP with 5% Cityscapes labeled data. The code will be public.

Problem

• Decoupling class and mask quality for pseudo-labels
• Correcting instance category labels dynamically
• Reducing noise impact in pixel-level mask pseudo-labels

Innovation

• Decoupled dual-threshold filtering for instance labels
• Dynamic instance category correction module
• Pixel-level mask uncertainty-aware re-weighting