当前位置: X-MOL 学术Pattern Recogn. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Vector of Locally and Adaptively Aggregated Descriptors for Image Feature Representation
Pattern Recognition ( IF 7.5 ) Pub Date : 2021-03-18 , DOI: 10.1016/j.patcog.2021.107952
Jian Zhang , Yunyin Cao , Qun Wu

VLAD (Vector of Locally Aggregated Descriptors) has been widely adopted in image representation. However, the VLAD algorithm seeks for the algebraic sum of the residue vectors between the descriptors and the centroid of cluster they belong to, and this could decrease the discriminative power of feature representations. To this end, this paper originally proposes a VLAAD (Vector of Locally and Adaptively Aggregated Descriptors) framework to adaptively assign a weight to each residue vector. First, we compute the weights using the magnitude of each residue vector, and encapsulate the weighted VLAD block into ResNet to form an end-to-end Weighted NetVLAD method. To further enhance the discriminative power of the features, we subsequently replace the magnitude-based weight computation with a gating scheme to achieve automatic weight estimation. The enhanced version is named as Gated NetVLAD method. The experimental results on CIFAR-10, MNIST Digits, Pittsburgh Google street view and ImageNet-Dog datasets demonstrate the promotion in classification accuracy and retrieval mAP using VLAAD against several state-of-the-art methods.



中文翻译:

用于图像特征表示的局部和自适应聚合描述符向量

VLAD(局部集合描述符的向量)已在图像表示中被广泛采用。但是,VLAD算法在描述符和它们所属的簇的质心之间寻找残差矢量的代数和,这可能会降低特征表示的判别力。为此,本文最初提出了一种VLAAD(局部和自适应聚集描述符的向量)框架,以自适应地为每个残差向量分配权重。首先,我们使用每个残差矢量的大小计算权重,并将加权的VLAD块封装到ResNet中,以形成端到端的加权NetVLAD方法。为了进一步增强特征的判别能力,我们随后用门控方案取代了基于幅度的权重计算,以实现自动权重估计。增强的版本称为Gated NetVLAD方法。在CIFAR-10,MNIST Digits,匹兹堡Google街景和ImageNet-Dog数据集上的实验结果表明,与几种最新方法相比,使用VLAAD可以提高分类准确性和检索mAP。

更新日期:2021-03-29
down
wechat
bug