SCALER: SAM-Enhanced Collaborative Learning for Label-Deficient Concealed Object Segmentation

TL;DR

SCALER introduces a collaborative framework that jointly optimizes a segmenter and SAM, improving concealed object segmentation under limited labels through reciprocal supervision and alternating training phases.

Contribution

The paper proposes a novel reciprocal supervision framework that integrates consistency constraints and SAM-based pseudo-labeling for enhanced label-deficient concealed object segmentation.

Findings

SCALER improves performance across eight semi- and weakly-supervised tasks.

Reciprocal supervision enhances both lightweight segmenters and foundation models.

Alternating optimization phases effectively leverage pseudo-labels and SAM robustness.

Abstract

Existing methods for label-deficient concealed object segmentation (LDCOS) either rely on consistency constraints or Segment Anything Model (SAM)-based pseudo-labeling. However, their performance remains limited due to the intrinsic concealment of targets and the scarcity of annotations. This study investigates two key questions: (1) Can consistency constraints and SAM-based supervision be jointly integrated to better exploit complementary information and enhance the segmenter? and (2) beyond that, can the segmenter in turn guide SAM through reciprocal supervision, enabling mutual improvement? To answer these questions, we present SCALER, a unified collaborative framework toward LDCOS that jointly optimizes a mean-teacher segmenter and a learnable SAM. SCALER operates in two alternating phases. In \textbf{Phase \uppercase\expandafter{\romannumeral1}}, the segmenter is optimized under fixed SAM supervision using entropy-based image-level and uncertainty-based pixel-level weighting to select reliable pseudo-label regions and emphasize harder examples. In \textbf{Phase \uppercase\expandafter{\romannumeral2}}, SAM is updated via augmentation invariance and noise resistance losses, leveraging its inherent robustness to perturbations. Experiments demonstrate that SCALER yields consistent performance gains across eight semi- and weakly-supervised COS tasks. The results further suggest that SCALER can serve as a general training paradigm to enhance both lightweight segmenters and large foundation models under label-scarce conditions. Code will be released.