Skip to main content
HM: Hybrid Masking for Few-Shot Segmentation
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
  • Seonghyeon Moon, Rutgers University, United States
  • Samuel S. Sohn, Rutgers University, United States
  • Honglu Zhou, Rutgers University, United States
  • Sejong Yoon, The College of New Jersey, United States
  • Vladimir Pavlovic, Rutgers University, United States
  • Muhammad Haris Khan, Mohamed bin Zayed University of Artificial Intelligence
  • Mubbasir Kapadia, Rutgers University, United States
Document Type
Conference Proceeding

We study few-shot semantic segmentation that aims to segment a target object from a query image when provided with a few annotated support images of the target class. Several recent methods resort to a feature masking (FM) technique to discard irrelevant feature activations which eventually facilitates the reliable prediction of segmentation mask. A fundamental limitation of FM is the inability to preserve the fine-grained spatial details that affect the accuracy of segmentation mask, especially for small target objects. In this paper, we develop a simple, effective, and efficient approach to enhance feature masking (FM). We dub the enhanced FM as hybrid masking (HM). Specifically, we compensate for the loss of fine-grained spatial details in FM technique by investigating and leveraging a complementary basic input masking method. Experiments have been conducted on three publicly available benchmarks with strong few-shot segmentation (FSS) baselines. We empirically show improved performance against the current state-of-the-art methods by visible margins across different benchmarks. Our code and trained models are available at: © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Publication Date
  • Frequency modulation,
  • Semantics,
  • Feature masking,
  • Few-shot learning,
  • Few-shot segmentation,
  • Fine grained,
  • Masking technique,
  • Query images,
  • Segmentation masks,
  • Semantic segmentation,
  • Shot segmentation,
  • Target object,
  • Computer Vision and Pattern Recognition (cs.CV)

IR Deposit conditions: non-described

Citation Information
S. Moon et al, "HM: Hybrid Masking for Few-Shot Segmentation", in Computer Vision (ECCV 2022), Lecture Notes in Computer Science, vol. 13680, pp. 506-523, Oct. 2022, doi:10.1007/978-3-031-20044-1_29