论文阅读 [TPAMI-2022] Knowledge-Guided Multi-Label Few-Shot Learning for General Image Recognition
论文搜索(studyai.com)
搜索论文: Knowledge-Guided Multi-Label Few-Shot Learning for General Image Recognition
搜索论文: http://www.studyai.com/search/whole-site/?q=Knowledge-Guided+Multi-Label+Few-Shot+Learning+for+General+Image+Recognition
关键字(Keywords)
Semantics; Task analysis; Training; Image recognition; Correlation; Neural networks; Proposals; Image recognition; multi-label learning; few-shot learning; knowledge graph; graph reasoning
机器学习; 机器视觉; 自然语言处理
图像分类; 多标签分类; 知识图谱; 小样本学习; 多标签学习; 递归神经网络
摘要(Abstract)
Recognizing multiple labels of an image is a practical yet challenging task, and remarkable progress has been achieved by searching for semantic regions and exploiting label dependencies.
识别一幅图像的多个标签是一项实际而富有挑战性的任务，通过搜索语义区域和利用标签依赖性已经取得了显著的进展。.
However, current works utilize RNN/LSTM to implicitly capture sequential region/label dependencies, which cannot fully explore mutual interactions among the semantic regions/labels and do not explicitly integrate label co-occurrences.
然而，目前的工作利用RNN/LSTM隐式地捕获顺序区域/标签依赖，这不能充分探索语义区域/标签之间的相互作用，也不能显式地集成标签共现。.
In addition, these works require large amounts of training samples for each category, and they are unable to generalize to novel categories with limited samples.
此外，这些工作需要为每个类别提供大量的训练样本，并且无法用有限的样本推广到新的类别。.
To address these issues, we propose a knowledge-guided graph routing (KGGR) framework, which unifies prior knowledge of statistical label correlations with deep neural networks.
为了解决这些问题，我们提出了一个知识引导图路由（KGGR）框架，该框架将统计标签相关性的先验知识与深度神经网络相结合。.
The framework exploits prior knowledge to guide adaptive information propagation among different categories to facilitate multi-label analysis and reduce the dependency of training samples.
该框架利用先验知识指导不同类别之间的自适应信息传播，以便于多标签分析，减少训练样本的依赖性。.
Specifically, it first builds a structured knowledge graph to correlate different labels based on statistical label co-occurrence.
具体来说，它首先构建一个结构化的知识图，根据统计标签共现来关联不同的标签。.
Then, it introduces the label semantics to guide learning semantic-specific features to initialize the graph, and it exploits a graph propagation network to explore graph node interactions, enabling learning contextualized image feature representations.
然后，它引入了标签语义来指导学习特定于语义的特征来初始化图，并利用图传播网络来探索图节点的交互，从而能够学习上下文化的图像特征表示。.
Moreover, we initialize each graph node with the classifier weights for the corresponding label and apply another propagation network to transfer node messages through the graph.
此外，我们使用相应标签的分类器权重初始化每个图节点，并应用另一个传播网络通过图传输节点消息。.
In this way, it can facilitate exploiting the information of correlated labels to help train better classifiers, especially for labels with limited training samples.
这样，它可以方便地利用相关标签的信息来帮助训练更好的分类器，尤其是对于训练样本有限的标签。.
We conduct extensive experiments on the traditional multi-label image recognition (MLR) and multi-label few-shot learning (ML-FSL) tasks and show that our KGGR framework outperforms the current state-of-the-art methods by sizable margins on the public benchmarks…
我们在传统的多标签图像识别（MLR）和多标签少镜头学习（ML-FSL）任务上进行了大量实验，结果表明，我们的KGGR框架在公共基准上有相当大的优势，优于当前最先进的方法。。.
作者(Authors)
[‘Tianshui Chen’, ‘Liang Lin’, ‘Riquan Chen’, ‘Xiaolu Hui’, ‘Hefeng Wu’]