Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/144270
Title: DeepEMD : few-shot image classification with differentiable Earth Mover’s Distance and structured classifiers
Authors: Zhang, Chi
Cai, Yujun
Lin, Guosheng
Shen, Chunhua
Keywords: Engineering::Computer science and engineering
Issue Date: 2020
Source: Zhang, C., Cai, Y., Lin, G., & Shen, C. (2020). DeepEMD : few-shot image classification with differentiable Earth Mover’s Distance and structured classifiers. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 12203-12213.
Project: AISG-RP-2018-003 
RG126/17 (S) 
RG28/18 (S) 
Conference: IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2020
Abstract: In this paper, we address the few-shot classification task from a new perspective of optimal matching between image regions. We adopt the Earth Mover’s Distance (EMD) as a metric to compute a structural distance between dense image representations to determine image relevance. The EMD generates the optimal matching flows between structural elements that have the minimum matching cost, which is used to represent the image distance for classification. To generate the important weights of elements in the EMD formulation, we design a cross-reference mechanism, which can effectively minimize the impact caused by the cluttered background and large intra-class appearance variations. To handle k-shot classification, we propose to learn a structured fully connected layer that can directly classify dense image representations with the EMD. Based on the implicit function theorem, the EMD can be inserted as a layer into the network for end-to-end training. We conduct comprehensive experiments to validate our algorithm and we set new state-of-the-art performance on four popular few-shot classification benchmarks, namely miniImageNet, tieredImageNet, Fewshot-CIFAR100 (FC100) and Caltech-UCSD Birds-200-2011 (CUB).
URI: https://hdl.handle.net/10356/144270
Schools: School of Computer Science and Engineering 
Rights: © 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Fulltext Permission: open
Fulltext Availability: With Fulltext
Appears in Collections:SCSE Conference Papers

Files in This Item:
File Description SizeFormat 
zhangchi 05957.pdf1.02 MBAdobe PDFThumbnail
View/Open

Page view(s)

282
Updated on May 21, 2024

Download(s) 20

322
Updated on May 21, 2024

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.