Person re-identification using adversarial haze attack and defense: A deep learning framework

Kanwal, Shansa; Shah, Jamal Hussain; Khan, Muhammad Attique; Nisa, Maryam; Kadry, Seifedine; Sharif, Muhammad; Yasmin, Mussarat; Maheswari, M.

doi:10.1016/j.compeleceng.2021.107542

Cited by 11 publications

(1 citation statement)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the ReID dataset was collected in clear weather without considering that the images would be affected by bad weather such as snow, rain and fog. In these adverse environments, the visibility of the images is very low, so Kanwal et al [12] addressed the problem of adversarial fog attack by the dark channel prior (DCP) method and used a fusion algorithm to fuse the handcrafted features with the features of the neural network to improve the effectiveness of ReID. Pang et al [13] proposed a novel Interference Suppression Model (ISM) to cope with the effect of severe weather on ReID.…”

Section: Introductionmentioning

confidence: 99%

Occluded Person Re-Identification by Multi-Granularity Generation Adversarial Network

et al. 2023

View full text Add to dashboard Cite

In order to address the problem that the detailed features of pedestrians are not prominent and the pedestrian pictures are obscured in unique environments in the process of person re-recognition, we propose a person re-recognition method with a multi-grain size generative adversarial network. Firstly, we use the generative adversarial network to recover the occluded pedestrian pictures; secondly, we improve the traditional multi-granularity network by adding an Efficient Channel Attention for Deep Convolutional Neural Networks (ECA-Net) on the coarse-grained branch to focus on the feature information in the pedestrian pictures and use the High-Resolution Net (HRNet) for pose estimation on the fine-grained branch to divide the pedestrian pictures into nine parts, to enhance the network's learning of more detailed features of pedestrians, and thus improve the accuracy of pedestrian re-recognition learning, which in turn improves the accuracy of person re-identification.INDEX TERMS Person re-identification, generative adversarial networks, random occlusion, attention mechanism.

show abstract

Section: Introductionmentioning

confidence: 99%