Search papers, labs, and topics across Lattice.
This study introduces Lite-Polyp Inductor (Lite-Pi), a novel framework that enhances lightweight models for polyp segmentation by leveraging foundation model induction. By generating FM-specific prototype representations and aligning them semantically through reconstruction-based supervision, Lite-Pi effectively improves generalization across diverse datasets while maintaining low computational costs. Extensive experiments reveal that Lite-Pi outperforms existing lightweight baselines, achieving superior segmentation accuracy with minimal resource requirements, making it a viable solution for real-time clinical applications.
Lightweight models can achieve superior polyp segmentation performance by effectively harnessing foundation model induction with minimal computational overhead.
Automated polyp segmentation in colonoscopy continues to pose challenges due to substantial appearance variations and indistinct polyp boundaries. Although emerging foundation models (FMs) such as DINOv2, SAM, and OneFormer, demonstrate remarkable generalization capabilities, their direct transfer to the polyp segmentation task and deployment in real-time clinical settings are difficult due to lack of large-scale labeled data and high computational demands. In addition, adopting multiple FMs together raises concerns, even though they encode complementary semantic and structural information. While lightweight models, including U-Net, PraNet and U-Net++, are computationally efficient, they often struggle to generalize across datasets due to limited representational capacity. To address this gap, we propose Lite-Polyp Inductor (Lite-Pi), a novel foundation model induction framework that significantly enhances lightweight polyp segmentation baselines. Our proposed framework generates FM-specific prototype representations and aligns them semantically with the corresponding foundation model priors through reconstruction-based supervision. Subsequently, transformer-based fusion is introduced to highlight the polyp relevant representations, including salient boundary information, while preserving complementary semantic cues. Extensive experiments across five polyp segmentation benchmark datasets demonstrate that Lite-蟺 significantly improves lightweight baselines, achieving superior generalization performance with minimal computational overhead and thereby, offering a practical solution for generalized polyp segmentation. Our code is available at GitHub. https://github.com/lostinrepo/Lite-Pi
