Back to Search
Start Over
Learning to Segment the Tail
- 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.
- Subjects :
- Scheme (programming language)
FOS: Computer and information sciences
Class (computer programming)
Forgetting
Computer science
business.industry
Computer Vision and Pattern Recognition (cs.CV)
Computer Science - Computer Vision and Pattern Recognition
Sample (statistics)
02 engineering and technology
010501 environmental sciences
01 natural sciences
Data modeling
Task (project management)
Visualization
Incremental learning
0202 electrical engineering, electronic engineering, information engineering
Task analysis
020201 artificial intelligence & image processing
Artificial intelligence
business
computer
0105 earth and related environmental sciences
computer.programming_language
Subjects
Details
- Database :
- OpenAIRE
- Journal :
- CVPR
- Accession number :
- edsair.doi.dedup.....27606b5e84d8f6b763231f52932e7d76
- Full Text :
- https://doi.org/10.48550/arxiv.2004.00900