Few-shot fine-grained classification with Spatial Attentive Comparison

The main goal of this paper is to propose a novel model, named Spatial Attentive Comparison Network (SACN), which is used to address a problem, termed few-shot fine-grained recognition (FSFG). FSFG is to recognize fine-grained examples with only a few samples, which is challenging for deep neural ne...

Full description

Saved in:
Bibliographic Details
Main Authors: Ruan, Xiaoqian, Lin, Guosheng, Long, Cheng, Lu, Shengli
Other Authors: School of Computer Science and Engineering
Format: Article
Language:English
Published: 2022
Subjects:
Online Access:https://hdl.handle.net/10356/160696
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-160696
record_format dspace
spelling sg-ntu-dr.10356-1606962022-08-01T04:05:29Z Few-shot fine-grained classification with Spatial Attentive Comparison Ruan, Xiaoqian Lin, Guosheng Long, Cheng Lu, Shengli School of Computer Science and Engineering Engineering::Computer science and engineering Few-Shot Learning Fine-Grained Classification The main goal of this paper is to propose a novel model, named Spatial Attentive Comparison Network (SACN), which is used to address a problem, termed few-shot fine-grained recognition (FSFG). FSFG is to recognize fine-grained examples with only a few samples, which is challenging for deep neural networks. SACN is made up of three modules, namely feature extraction module, selective-comparison similarity module (SCSM), and classification module: feature extraction module extracts the distinctive information into feature maps, SCSM is used to fuse the features of support set with those of the query set based on selective comparison. Considering the noisy background and tiny differences between different categories, we apply SCSM to fuse these features by arranging different weights pixel by pixel, and all these weights are learned automatically. Moreover, we apply pyramid structure to enrich the features. By conducting comprehensive experiments on three fine-grained datasets, namely CUB-200-2011 (CUB Birds), Stanford Dogs Dataset, and Stanford Cars Dataset, we demonstrate that the proposed method achieves superior performance over the competing baselines. Ministry of Education (MOE) This work is partly supported by MOE Tier-1 Singapore research grants: RG28/18 (S) and RG22/19 (S). 2022-08-01T04:05:29Z 2022-08-01T04:05:29Z 2021 Journal Article Ruan, X., Lin, G., Long, C. & Lu, S. (2021). Few-shot fine-grained classification with Spatial Attentive Comparison. Knowledge-Based Systems, 218, 106840-. https://dx.doi.org/10.1016/j.knosys.2021.106840 0950-7051 https://hdl.handle.net/10356/160696 10.1016/j.knosys.2021.106840 2-s2.0-85101112161 218 106840 en RG28/18 (S) RG22/19 (S) Knowledge-Based Systems © 2021 Elsevier B.V. All rights reserved.
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Engineering::Computer science and engineering
Few-Shot Learning
Fine-Grained Classification
spellingShingle Engineering::Computer science and engineering
Few-Shot Learning
Fine-Grained Classification
Ruan, Xiaoqian
Lin, Guosheng
Long, Cheng
Lu, Shengli
Few-shot fine-grained classification with Spatial Attentive Comparison
description The main goal of this paper is to propose a novel model, named Spatial Attentive Comparison Network (SACN), which is used to address a problem, termed few-shot fine-grained recognition (FSFG). FSFG is to recognize fine-grained examples with only a few samples, which is challenging for deep neural networks. SACN is made up of three modules, namely feature extraction module, selective-comparison similarity module (SCSM), and classification module: feature extraction module extracts the distinctive information into feature maps, SCSM is used to fuse the features of support set with those of the query set based on selective comparison. Considering the noisy background and tiny differences between different categories, we apply SCSM to fuse these features by arranging different weights pixel by pixel, and all these weights are learned automatically. Moreover, we apply pyramid structure to enrich the features. By conducting comprehensive experiments on three fine-grained datasets, namely CUB-200-2011 (CUB Birds), Stanford Dogs Dataset, and Stanford Cars Dataset, we demonstrate that the proposed method achieves superior performance over the competing baselines.
author2 School of Computer Science and Engineering
author_facet School of Computer Science and Engineering
Ruan, Xiaoqian
Lin, Guosheng
Long, Cheng
Lu, Shengli
format Article
author Ruan, Xiaoqian
Lin, Guosheng
Long, Cheng
Lu, Shengli
author_sort Ruan, Xiaoqian
title Few-shot fine-grained classification with Spatial Attentive Comparison
title_short Few-shot fine-grained classification with Spatial Attentive Comparison
title_full Few-shot fine-grained classification with Spatial Attentive Comparison
title_fullStr Few-shot fine-grained classification with Spatial Attentive Comparison
title_full_unstemmed Few-shot fine-grained classification with Spatial Attentive Comparison
title_sort few-shot fine-grained classification with spatial attentive comparison
publishDate 2022
url https://hdl.handle.net/10356/160696
_version_ 1743119586748792832