region proposal方法不是随机或密集地生成多尺度图像块,而是在处理图像对象时引入了一定程度的人肉规则以过滤掉明显无用的局部image,这种方法效率更高,内存需求更少。可以使用无监督对象检测器生成区域建议,例如选择性搜索[74] 和边缘框 [75]。除了使用物体检测器,Xie 等人[76] 介绍了一种手动对象检测方法,其中建议...
content based image retrieval (CBIR), is a long-established research area, and more efficient and accurate methods are needed for real time retrieval. Artificial intelligence has made progress in CBIR and has significantly facilitated the process of intelligent search. In this survey we organize and...
In this paper, we present a comprehensive deep hashing survey for the task of image retrieval with multiple labels, categorizing the methods according to how the input images are treated: pointwise, pairwise, tripletwise and listwise, as well as their relationships. In addition, we present ...
细粒度识别是FGIA研究最多的领域,因为识别是大多数视觉系统的基本能力,因此值得长期持续研究。 细粒度检索(Fine-Grained Retrieval):根据查询图像的类型将细粒度检索方法分为两组 基于内容的细粒度图像检索 基于草图的细粒度图片检索。 与细粒度识别相比,细粒度检索是近年来FGIA的一个新兴领域,越来越受到学术界和工业界...
A Survey on Transferability of Adversarial Examples across Deep Neural Networksarxiv.org/abs/2310.17626 摘要 深度神经网络(DNN)的出现彻底改变了各个领域,使图像识别、自然语言处理和科学问题解决等复杂任务的解决成为可能。然而,这一进展也暴露了一个令人担忧的脆弱性:对抗性示例。这些精心制作的输入,人类无法...
Term-weighting approaches in automatic text retrieval Inf. Process. Manage. (1988) RuiY. et al. Image retrieval: Current techniques, promising directions, and open issues J. Vis. Commun. Image Represent. (1999) LiuW. et al. A survey of deep neural network architectures and their applications...
image.png raw data 往往包含较大噪声 bounding box 一般采用检测和跟踪算法获得 annotation 训练数据标定是最耗费人力成本的步骤,也是有监督学习不可或缺的 training 模型的建立,包括特征抽取方式,度量学习方式,损失函数等。该阶段是研究的重点 retrieval 推理阶段,一般通过gallery中相似度进行排序后,评估性能。
To better illustrate the difference of the key contributions in the past and this survey, Table1summarises the main deep face recognition surveys. The analysis presented by Wang et al. [46] is arguably the most comprehensive survey yet in the field. It provides a holistic overview of the bro...
Supervised deep hashing for scalable face image retrieval Pattern Recognit. (2018) J. Wang et al. A survey on learning to hash IEEE Trans. Pattern Anal. Mach. Intell. (2018) P. Li et al. Hashing algorithms for large-scale learning NIPS (2011) Y. Gong et al. Iterative quantization: a...
A Survey on Deep Learning: Algorithms, Techniques, and Applications 成员: 惠州学院大二在校生-庄思杰 惠州学院大三在校生-邹旭智 导师:罗除 Abstract 随着深度学习逐渐成为该领域的领导者,机器学习领域正在见证它的黄金时代。深度学习使用多层来表示数据的抽象,以建立计算模型。一些关键的使能深度学习算法,如生成对抗...