Through the Looking Glass: A Dual Perspective on Weakly-Supervised Few-Shot Segmentation.
basic_science · Level V
Where this comes from
- Record sourced from PubMed, PMID 42406674.
- Also identified by DOI 10.1109/TIP.2026.3708050.
- No licence information is recorded for this record.
- Because redistribution is not established, this page shows the abstract only. Follow the links below for the full text.
Abstract
Meta-learning aims to uniformly sample homologous support-query pairs, characterized by the same categories and similar attributes, and extract useful inductive biases through identical network architectures. However, this identical network design results in over-semantic homogenization. To address this, we propose a novel homologous but heterogeneous network. By treating support-query pairs as dual perspectives, we introduce heterogeneous visual aggregation (HA) modules to enhance complementarity while preserving semantic commonality. To further reduce semantic noise and amplify the uniqueness of heterogeneous semantics, we design a heterogeneous transfer (HT) module. Finally, we propose heterogeneous CLIP (HC) textual information to enhance the generalization capability of multimodal models. In the weakly-supervised few-shot semantic segmentation (WFSS) task, with only 1/24 of the parameters of existing state-of-the-art models, TLG achieves a 13.2% improvement on Pascal-5<sup>i</sup> and a 7.9% improvement on COCO-20<sup>i</sup>. To the best of our knowledge, TLG is also the first weakly-supervised (image-level) model that outperforms fully supervised (pixel-level) models under the same backbone architectures. The code is available at https://github.com/jarch-ma/TLG.