论文阅读 [TPAMI-2022] Hierarchical Deep Click Feature Prediction for Fine-Grained Image Recognition
论文搜索(studyai.com)
搜索论文: Hierarchical Deep Click Feature Prediction for Fine-Grained Image Recognition
搜索论文: http://www.studyai.com/search/whole-site/?q=Hierarchical+Deep+Click+Feature+Prediction+for+Fine-Grained+Image+Recognition
关键字(Keywords)
Visualization; Feature extraction; Image recognition; Semantics; Predictive models; Vocabulary; Task analysis; Click prediction; hierarchical model; word embedding; deep neural network; transfer learning
机器学习; 机器视觉; 自然语言处理
图像分类; 单样本学习; 词嵌入; 细粒度视觉; 迁移学习
摘要(Abstract)
The click feature of an image, defined as the user click frequency vector of the image on a predefined word vocabulary, is known to effectively reduce the semantic gap for fine-grained image recognition.
图像的点击特征定义为用户在预定义词汇表上点击图像的频率向量,可以有效地缩小细粒度图像识别的语义差距。.
Unfortunately, user click frequency data are usually absent in practice.
不幸的是,用户点击频率数据在实践中通常是不存在的。.
It remains challenging to predict the click feature from the visual feature, because the user click frequency vector of an image is always noisy and sparse.
从视觉特征预测点击特征仍然是一个挑战,因为图像的用户点击频率向量总是噪声和稀疏的。.
In this paper, we devise a Hierarchical Deep Word Embedding (HDWE) model by integrating sparse constraints and an improved RELU operator to address click feature prediction from visual features.
在本文中,我们通过集成稀疏约束和改进的RELU算子,设计了一个层次化深层单词嵌入(HDWE)模型,以解决视觉特征的点击特征预测问题。.
HDWE is a coarse-to-fine click feature predictor that is learned with the help of an auxiliary image dataset containing click information.
HDWE是一个从粗到细的点击特征预测器,它是在包含点击信息的辅助图像数据集的帮助下学习的。.
It can therefore discover the hierarchy of word semantics.
因此,它可以发现单词语义的层次结构。.
We evaluate HDWE on three dog and one bird image datasets, in which Clickture-Dog and Clickture-Bird are utilized as auxiliary datasets to provide click data, respectively.
我们在三只狗和一只鸟的图像数据集上评估HDWE,其中Clickture dog和Clickture bird分别作为辅助数据集提供点击数据。.
Our empirical studies show that HDWE has 1) higher recognition accuracy, 2) a larger compression ratio, and 3) good one-shot learning ability and scalability to unseen categories…
我们的实证研究表明,HDWE具有1)更高的识别精度,2)更大的压缩比,以及3)良好的一次性学习能力和对未知类别的可扩展性。。.
作者(Authors)
[‘Jun Yu’, ‘Min Tan’, ‘Hongyuan Zhang’, ‘Yong Rui’, ‘Dacheng Tao’]