论文标题
MHSA-NET：闭塞人重新识别的多头自我发项网络
MHSA-Net: Multi-Head Self-Attention Network for Occluded Person Re-Identification
论文作者
论文摘要
本文介绍了一种新颖的人重新识别模型，称为多头自我发项网络（MHSA-NET），以修剪不重要的信息并从人图像中捕获关键的本地信息。 MHSA-NET包含两个主要的新颖组成部分：多头自我注意力分支（MHSAB）和注意竞争机制（ACM）。 MHSAB适应捕获关键的本地人信息，然后为匹配者提供有效的图像的多样性嵌入。 ACM进一步有助于滤除注意力噪声和非关键信息。通过广泛的消融研究，我们验证了多头自我注意力分支（MHSAB）和注意竞争机制（ACM）都有助于MHSA-NET的性能提高。我们的MHSA-NET在标准和被遮挡的人重新ID任务中实现了竞争性能。
This paper presents a novel person re-identification model, named Multi-Head Self-Attention Network (MHSA-Net), to prune unimportant information and capture key local information from person images. MHSA-Net contains two main novel components: Multi-Head Self-Attention Branch (MHSAB) and Attention Competition Mechanism (ACM). The MHSAB adaptively captures key local person information, and then produces effective diversity embeddings of an image for the person matching. The ACM further helps filter out attention noise and non-key information. Through extensive ablation studies, we verified that the Multi-Head Self-Attention Branch (MHSAB) and Attention Competition Mechanism (ACM) both contribute to the performance improvement of the MHSA-Net. Our MHSA-Net achieves competitive performance in the standard and occluded person Re-ID tasks.
