Paper published in a journal (Scientific congresses, symposiums and conference proceedings)
MULTI LABEL IMAGE CLASSIFICATION USING ADAPTIVE GRAPH CONVOLUTIONAL NETWORKS (ML-AGCN)
Singh, Inder Pal; Ghorbel, Enjie; Oyedotun, Oyebade et al.
2022In IEEE International Conference on Image Processing
Peer reviewed
 

Files


Full Text
ML_AGCN_Camera_Ready_ICIP_2022_InderPalSingh.pdf
Author postprint (623.97 kB)
Download

All documents in ORBilu are protected by a user license.

Send to



Details



Keywords :
Computer Vision; Machine Learning; Multi-label Image Classification; Graph Convolutional Networks; Deep Learning; Image Processing
Abstract :
[en] In this paper, a novel graph-based approach for multi-label image classification called Multi-Label Adaptive Graph Convolutional Network (ML-AGCN) is introduced. Graph-based methods have shown great potential in the field of multi-label classification. However, these approaches heuristically fix the graph topology for modeling label dependencies, which might be not optimal. To handle that, we propose to learn the topology in an end-to-end manner. Specifically, we incorporate an attention-based mechanism for estimating the pairwise importance between graph nodes and a similarity-based mechanism for conserving the feature similarity between different nodes. This offers a more flexible way for adaptively modeling the graph. Experimental results are reported on two well-known datasets, namely, MS-COCO and VG-500. Results show that ML-AGCN outperforms state-of-the-art methods while reducing the number of model parameters.
Research center :
SnT - Interdisciplinary Centre for Security, Reliability and Trust, University of Luxembourg
Disciplines :
Computer science
Author, co-author :
Singh, Inder Pal  ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > CVI2
Ghorbel, Enjie  ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > CVI2
Oyedotun, Oyebade ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > CVI2
Aouada, Djamila  ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > CVI2
External co-authors :
no
Language :
English
Title :
MULTI LABEL IMAGE CLASSIFICATION USING ADAPTIVE GRAPH CONVOLUTIONAL NETWORKS (ML-AGCN)
Publication date :
2022
Event name :
2022 IEEE International Conference on Image Processing (ICIP)
Event organizer :
IEEE
Event place :
Bordeaux, France
Event date :
15-10-2022 to 19-10-2022
Audience :
International
Journal title :
IEEE International Conference on Image Processing
Peer reviewed :
Peer reviewed
Focus Area :
Security, Reliability and Trust
FnR Project :
FNR14755859 - Multi-modal Fusion Of Electro-optical Sensors For Spacecraft Pose Estimation Towards Autonomous In-orbit Operations, 2020 (01/01/2021-31/12/2023) - Djamila Aouada
Name of the research project :
MEET-A
Funders :
FNR - Fonds National de la Recherche [LU]
Available on ORBilu :
since 25 July 2022

Statistics


Number of views
160 (17 by Unilu)
Number of downloads
280 (34 by Unilu)

Bibliography


Similar publications



Contact ORBilu