Meta-RCNN: Meta learning for few-shot object detection

Author

Xiongwei WU
Doyen SAHOO
Steven HOI, Singapore Management UniversityFollow

Publication Type

Conference Proceeding Article

Version

acceptedVersion

Publication Date

10-2020

Abstract

Despite significant advances in deep learning based object detection in recent years, training effective detectors in a small data regime remains an open challenge. This is very important since labelling training data for object detection is often very expensive and time-consuming. In this paper, we investigate the problem of few-shot object detection, where a detector has access to only limited amounts of annotated data. Based on the meta-learning principle, we propose a new meta-learning framework for object detection named "Meta-RCNN", which learns the ability to perform few-shot detection via meta-learning. Specifically, Meta-RCNN learns an object detector in an episodic learning paradigm on the (meta) training data. This learning scheme helps acquire a prior which enables Meta-RCNN to do few-shot detection on novel tasks. Built on top of the popular Faster RCNN detector, in Meta-RCNN, both the Region Proposal Network (RPN) and the object classification branch are meta-learned. The meta-trained RPN learns to provide class-specific proposals, while the object classifier learns to do few-shot classification. The novel loss objectives and learning strategy of Meta-RCNN can be trained in an end-to-end manner. We demonstrate the effectiveness of Meta-RCNN in few-shot detection on three datasets (Pascal-VOC, ImageNet-LOC and MSCOCO) with promising results.

Keywords

Object Detection, Deep Learning, Meta Learning

Discipline

Databases and Information Systems | Data Science

Research Areas

Data Science and Engineering

Areas of Excellence

Digital transformation

Publication

MM '20: Proceedings of the 28th ACM International Conference on Multimedia, Seattle, WA, USA, October 12-16

First Page

1679

Last Page

1687

Identifier

10.1145/3394171.3413832

Publisher

ACM

City or Country

New York

Embargo Period

2-10-2025

Citation

WU, Xiongwei; SAHOO, Doyen; and HOI, Steven. Meta-RCNN: Meta learning for few-shot object detection. (2020). MM '20: Proceedings of the 28th ACM International Conference on Multimedia, Seattle, WA, USA, October 12-16. 1679-1687.
Available at: https://ink.library.smu.edu.sg/sis_research/10097

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.

Additional URL

https://doi.org/10.1145/3394171.3413832