mlsanityData to 7 Oct 2026

Background Clustering Pre-training for Few-shot Segmentation

arXiv 2312.03322 ↗First posted 6 Dec 2023cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Recent few-shot segmentation (FSS) methods introduce an extra pre-training stage before meta-training to obtain a stronger backbone, which has become a standard step in few-shot learning. Despite the effectiveness, current pre-training scheme suffers from the merged background problem: only base classes are labelled as foregrounds, making it hard to distinguish between novel classes and actual background. In this paper, we propose a new pre-training scheme for FSS via decoupling the novel classes from background, called Background Clustering Pre-Training (BCPT). Specifically, we adopt online clustering to the pixel embeddings of merged background to explore the underlying semantic structures, bridging the gap between pre-training and adaptation to novel classes. Given the clustering results, we further propose the background mining loss and leverage base classes to guide the clustering process, improving the quality and stability of clustering results. Experiments on PASCAL-5i and COCO-20i show that BCPT yields advanced performance. Code will be available.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 21 of its 26 references that are in the map. First posted on arXiv 6 Dec 2023.

Builds on

21 of 26 references.

References · 21

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  3. 2015Unsupervised Visual Representation Learning by Context Predictioncited by 1,062
  4. 2018Deep Clustering for Unsupervised Learning of Visual Featurescited by 1,020
  5. 2017One-Shot Learning for Semantic Segmentationcited by 337
  6. 2018Few-Shot Learning via Embedding Adaptation with Set-to-Set Functionscited by 293
  7. 2020Prior Guided Feature Enrichment Network for Few-Shot Segmentationcited by 271
  8. 2020Prototype Mixture Models for Few-shot Semantic Segmentationcited by 190
  9. 2019Feature Weighting and Boosting for Few-Shot Segmentationcited by 189
  10. 2020Part-aware Prototype Network for Few-shot Semantic Segmentationcited by 162
  11. 2021Few-Shot Segmentation via Cycle-Consistent Transformercited by 133
  12. 2022Learning What Not to Segment: A New Perspective on Few-Shot Segmentationcited by 120
9 more
  1. 2021Simpler is Better: Few-shot Semantic Segmentation with Classifier Weight Transformercited by 116
  2. 2020Online Deep Clustering for Unsupervised Representation Learningcited by 99
  3. 2022Self-Support Few-Shot Semantic Segmentationcited by 96
  4. 2021Mining Latent Classes for Few-shot Segmentationcited by 69
  5. 2022Dense Cross-Query-and-Support Attention Weighted Mask Aggregation for Few-Shot Segmentationcited by 65
  6. 2022Singular Value Fine-tuning: Few-shot Segmentation requires Few-parameters Fine-tuningcited by 50
  7. 2022Dynamic Prototype Convolution Network for Few-Shot Semantic Segmentationcited by 43
  8. 2022MSANet: Multi-Similarity and Attention Guidance for Boosting Few-Shot Segmentationcited by 21
  9. 2021Cost Aggregation Is All You Need for Few-Shot Segmentationcited by 16

Built on by

Not cited yet.