SNN-PAR: Energy Efficient Pedestrian Attribute Recognition via Spiking Neural Networks

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Wang, Haiyang, Zhu, Qian, She, Mowen, Li, Yabo, Song, Haoyu, Xu, Minghe, Wang, Xiao
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913540175036416
author Wang, Haiyang
Zhu, Qian
She, Mowen
Li, Yabo
Song, Haoyu
Xu, Minghe
Wang, Xiao
author_facet Wang, Haiyang
Zhu, Qian
She, Mowen
Li, Yabo
Song, Haoyu
Xu, Minghe
Wang, Xiao
contents Artificial neural network based Pedestrian Attribute Recognition (PAR) has been widely studied in recent years, despite many progresses, however, the energy consumption is still high. To address this issue, in this paper, we propose a Spiking Neural Network (SNN) based framework for energy-efficient attribute recognition. Specifically, we first adopt a spiking tokenizer module to transform the given pedestrian image into spiking feature representations. Then, the output will be fed into the spiking Transformer backbone networks for energy-efficient feature extraction. We feed the enhanced spiking features into a set of feed-forward networks for pedestrian attribute recognition. In addition to the widely used binary cross-entropy loss function, we also exploit knowledge distillation from the artificial neural network to the spiking Transformer network for more accurate attribute recognition. Extensive experiments on three widely used PAR benchmark datasets fully validated the effectiveness of our proposed SNN-PAR framework. The source code of this paper is released on \url{https://github.com/Event-AHU/OpenPAR}.
format Preprint
id arxiv_https___arxiv_org_abs_2410_07857
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle SNN-PAR: Energy Efficient Pedestrian Attribute Recognition via Spiking Neural Networks
Wang, Haiyang
Zhu, Qian
She, Mowen
Li, Yabo
Song, Haoyu
Xu, Minghe
Wang, Xiao
Computer Vision and Pattern Recognition
Artificial Intelligence
Neural and Evolutionary Computing
Artificial neural network based Pedestrian Attribute Recognition (PAR) has been widely studied in recent years, despite many progresses, however, the energy consumption is still high. To address this issue, in this paper, we propose a Spiking Neural Network (SNN) based framework for energy-efficient attribute recognition. Specifically, we first adopt a spiking tokenizer module to transform the given pedestrian image into spiking feature representations. Then, the output will be fed into the spiking Transformer backbone networks for energy-efficient feature extraction. We feed the enhanced spiking features into a set of feed-forward networks for pedestrian attribute recognition. In addition to the widely used binary cross-entropy loss function, we also exploit knowledge distillation from the artificial neural network to the spiking Transformer network for more accurate attribute recognition. Extensive experiments on three widely used PAR benchmark datasets fully validated the effectiveness of our proposed SNN-PAR framework. The source code of this paper is released on \url{https://github.com/Event-AHU/OpenPAR}.
title SNN-PAR: Energy Efficient Pedestrian Attribute Recognition via Spiking Neural Networks
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Neural and Evolutionary Computing
url https://arxiv.org/abs/2410.07857