Amazing technological breakthrough possible @S-Logix

Office Address

  • #5, First Floor, 4th Street Dr. Subbarayan Nagar Kodambakkam, Chennai-600 024 Landmark : Samiyar Madam
  • +91- 81240 01111

Social List

Prior guided feature enrichment network for few-shot segmentation - 2020

Prior Guided Feature Enrichment Network For Few-Shot Segmentation

Research Area:  Machine Learning


State-of-the-art semantic segmentation methods require sufficient labeled data to achieve good results and hardly work on unseen classes without fine-tuning. Few-shot segmentation is thus proposed to tackle this problem by learning a model that quickly adapts to new classes with a few labeled support samples. Theses frameworks still face the challenge of generalization ability reduction on unseen classes due to inappropriate use of high-level semantic information of training classes and spatial inconsistency between query and support targets. To alleviate these issues, we propose the Prior Guided Feature Enrichment Network (PFENet). It consists of novel designs of (1) a training-free prior mask generation method that not only retains generalization power but also improves model performance and (2) Feature Enrichment Module (FEM) that overcomes spatial inconsistency by adaptively enriching query features with support features and prior masks. Extensive experiments on PASCAL-5ii and COCO prove that the proposed prior generation method and FEM both improve the baseline method significantly. Our PFENet also outperforms state-of-the-art methods by a large margin without efficiency loss. It is surprising that our model even generalizes to cases without labeled support samples.


Author(s) Name:  Zhuotao Tian,Hengshuang Zhao,Michelle Shu,Zhicheng Yang, Ruiyu Li,Jiaya Jia,

Journal name:  IEEE Transactions on Pattern Analysis and Machine Intelligence

Conferrence name:  

Publisher name:  IEEE

DOI:  10.1109/TPAMI.2020.3013717

Volume Information:  Feb. 2022, pp. 1050-1065, vol. 44