Lightweight Attribute Localizing Models for Pedestrian Attribute Recognition

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Jha, Ashish, Ermilov, Dimitrii, Sobolev, Konstantin, Phan, Anh Huy, Ahmadi-Asl, Salman, Ahmed, Naveed, Junejo, Imran, Aghbari, Zaher AL, Baker, Thar, Khedr, Ahmed Mohamed, Cichocki, Andrzej
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918125335740416
author Jha, Ashish
Ermilov, Dimitrii
Sobolev, Konstantin
Phan, Anh Huy
Ahmadi-Asl, Salman
Ahmed, Naveed
Junejo, Imran
Aghbari, Zaher AL
Baker, Thar
Khedr, Ahmed Mohamed
Cichocki, Andrzej
author_facet Jha, Ashish
Ermilov, Dimitrii
Sobolev, Konstantin
Phan, Anh Huy
Ahmadi-Asl, Salman
Ahmed, Naveed
Junejo, Imran
Aghbari, Zaher AL
Baker, Thar
Khedr, Ahmed Mohamed
Cichocki, Andrzej
contents Pedestrian Attribute Recognition (PAR) focuses on identifying various attributes in pedestrian images, with key applications in person retrieval, suspect re-identification, and soft biometrics. However, Deep Neural Networks (DNNs) for PAR often suffer from over-parameterization and high computational complexity, making them unsuitable for resource-constrained devices. Traditional tensor-based compression methods typically factorize layers without adequately preserving the gradient direction during compression, leading to inefficient compression and a significant accuracy loss. In this work, we propose a novel approach for determining the optimal ranks of low-rank layers, ensuring that the gradient direction of the compressed model closely aligns with that of the original model. This means that the compressed model effectively preserves the update direction of the full model, enabling more efficient compression for PAR tasks. The proposed procedure optimizes the compression ranks for each layer within the ALM model, followed by compression using CPD-EPC or truncated SVD. This results in a reduction in model complexity while maintaining high performance.
format Preprint
id arxiv_https___arxiv_org_abs_2306_09822
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Lightweight Attribute Localizing Models for Pedestrian Attribute Recognition
Jha, Ashish
Ermilov, Dimitrii
Sobolev, Konstantin
Phan, Anh Huy
Ahmadi-Asl, Salman
Ahmed, Naveed
Junejo, Imran
Aghbari, Zaher AL
Baker, Thar
Khedr, Ahmed Mohamed
Cichocki, Andrzej
Computer Vision and Pattern Recognition
Numerical Analysis
Pedestrian Attribute Recognition (PAR) focuses on identifying various attributes in pedestrian images, with key applications in person retrieval, suspect re-identification, and soft biometrics. However, Deep Neural Networks (DNNs) for PAR often suffer from over-parameterization and high computational complexity, making them unsuitable for resource-constrained devices. Traditional tensor-based compression methods typically factorize layers without adequately preserving the gradient direction during compression, leading to inefficient compression and a significant accuracy loss. In this work, we propose a novel approach for determining the optimal ranks of low-rank layers, ensuring that the gradient direction of the compressed model closely aligns with that of the original model. This means that the compressed model effectively preserves the update direction of the full model, enabling more efficient compression for PAR tasks. The proposed procedure optimizes the compression ranks for each layer within the ALM model, followed by compression using CPD-EPC or truncated SVD. This results in a reduction in model complexity while maintaining high performance.
title Lightweight Attribute Localizing Models for Pedestrian Attribute Recognition
topic Computer Vision and Pattern Recognition
Numerical Analysis
url https://arxiv.org/abs/2306.09822