Back to Search Start Over

Learning to Segment the Tail

Authors :
Xinting Hu
Yi Jiang
Chunyan Miao
Jingyuan Chen
Kaihua Tang
Hanwang Zhang
Source :
CVPR
Publication Year :
2020
Publisher :
arXiv, 2020.

Abstract

Real-world visual recognition requires handling the extreme sample imbalance in large-scale long-tailed data. We propose a "divide&conquer" strategy for the challenging LVIS task: divide the whole data into balanced parts and then apply incremental learning to conquer each one. This derives a novel learning paradigm: class-incremental few-shot learning, which is especially effective for the challenge evolving over time: 1) the class imbalance among the old-class knowledge review and 2) the few-shot data in new-class learning. We call our approach Learning to Segment the Tail (LST). In particular, we design an instance-level balanced replay scheme, which is a memory-efficient approximation to balance the instance-level samples from the old-class images. We also propose to use a meta-module for new-class learning, where the module parameters are shared across incremental phases, gaining the learning-to-learn knowledge incrementally, from the data-rich head to the data-poor tail. We empirically show that: at the expense of a little sacrifice of head-class forgetting, we can gain a significant 8.3% AP improvement for the tail classes with less than 10 instances, achieving an overall 2.0% AP boost for the whole 1,230 classes.

Details

Database :
OpenAIRE
Journal :
CVPR
Accession number :
edsair.doi.dedup.....27606b5e84d8f6b763231f52932e7d76
Full Text :
https://doi.org/10.48550/arxiv.2004.00900