Publication Type
Conference Proceeding Article
Version
acceptedVersion
Publication Date
6-2023
Abstract
Few-shot open-set recognition (FSOR) is a challenging task of great practical value. It aims to categorize a sample to one of the pre-defined, closed-set classes illustrated by few examples while being able to reject the sample from unknown classes. In this work, we approach the FSOR task by proposing a novel energy-based hybrid model. The model is composed of two branches, where a classification branch learns a metric to classify a sample to one of closedset classes and the energy branch explicitly estimates the open-set probability. To achieve holistic detection of openset samples, our model leverages both class-wise and pixelwise features to learn a glocal energy-based score, in which a global energy score is learned using the class-wise features, while a local energy score is learned using the pixelwise features. The model is enforced to assign large energy scores to samples that are deviated from the few-shot examples in either the class-wise features or the pixel-wise features, and to assign small energy scores otherwise. Experiments on three standard FSOR datasets show the superior performance of our model.
Discipline
Databases and Information Systems | Theory and Algorithms
Research Areas
Data Science and Engineering
Publication
2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR): June 18-22, Vancouver: Proceedings
First Page
7507
Last Page
7516
Publisher
IEEE Computer Society
City or Country
Los Alamitos
Citation
WANG, Haoyu; PANG, Guansong; WANG, Peng; ZHANG, Lei; WEI, Wei; and ZHANG, Yanning.
Glocal energy-based learning for few-shot open-set recognition. (2023). 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR): June 18-22, Vancouver: Proceedings. 7507-7516.
Available at: https://ink.library.smu.edu.sg/sis_research/8005
Copyright Owner and License
Authors
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.