ICCV · 2021

FASA: Feature Augmentation and Sampling Adaptation for Long-Tailed Instance Segmentation

Yuhang Zang, Chen Huang, Chen Change Loy

Key takeaway

FASA addresses rare-class data scarcity by generating virtual features from observed class statistics and adapting their sampling to the model's training losses. arXiv abstract · v2

Abstract

Recent methods for long-tailed instance segmentation still struggle on rare object classes with few training data. We propose a simple yet effective method, Feature Augmentation and Sampling Adaptation (FASA), that addresses the data scarcity issue by augmenting the feature space especially for rare classes. Both the Feature Augmentation (FA) and feature sampling components are adaptive to the actual training status -- FA is informed by the feature mean and variance of observed real samples from past iterations, and we sample the generated virtual features in a loss-adapted manner to avoid over-fitting. FASA does not require any elaborate loss design, and removes the need for inter-class transfer learning that often involves large cost and manually-defined head/tail class groups. We show FASA is a fast, generic method that can be easily plugged into standard or long-tailed segmentation frameworks, with consistent performance gains and little added cost. FASA is also applicable to other tasks like long-tailed classification with state-of-the-art performance.

Author abstract · arXiv abstract · v2

Publication

IEEE International Conference on Computer Vision (ICCV), 2021

Paper and resources

Research topics

FASA · Long-tailed instance segmentation · Feature augmentation · Adaptive sampling · Rare-class recognition · Virtual features · Class imbalance · Long-tailed classification

Research problem and approach

Long-tailed instance segmentation lacks enough examples for rare categories. FASA augments feature space using running means and variances, then adjusts virtual-feature sampling to avoid overfitting without manual head/tail partitions or an elaborate new loss. arXiv abstract · v2

Main contributions

  • Combines adaptive feature augmentation with loss-adapted sampling of virtual examples. arXiv abstract · v2
  • Provides a generic module for standard and long-tailed segmentation frameworks and extends it to classification. arXiv abstract · v2

Method comparison

ApproachKey difference
Inter-class transfer for rare categoriesCan require costly transfer procedures and manually defined head/tail class groups.
FASAGenerates features from observed class statistics and samples them adaptively using training losses.

arXiv abstract · v2

Selected results

Cite this paper

Yuhang Zang, Chen Huang, Chen Change Loy. FASA: Feature Augmentation and Sampling Adaptation for Long-Tailed Instance Segmentation. Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2021, pp. 3457–3466.

@inproceedings{arxiv210212867,
  title     = {{FASA: Feature Augmentation and Sampling Adaptation for Long-Tailed Instance Segmentation}},
  author    = {Yuhang Zang and Chen Huang and Chen Change Loy},
  booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
  month     = {October},
  year      = {2021},
  pages     = {3457--3466},
  url       = {https://openaccess.thecvf.com/content/ICCV2021/html/Zang_FASA_Feature_Augmentation_and_Sampling_Adaptation_for_Long-Tailed_Instance_Segmentation_ICCV_2021_paper.html}
}