ISIA Food-500：一个基于层叠全局-局部注意力网络的大规模食物识别数据集
ISIA Food-500: A Dataset for Large-Scale Food Recognition via Stacked Global-Local Attention Network
摘要：
食物识别因其在饮食管理、自助餐厅等方面的广泛应用而受到多媒体领域的广泛关注。大规模食品图像本体是开发先进的大规模食品识别算法的迫切需要，以及为这些算法提供基准数据集。为了鼓励食品识别的进一步发展，我们引入了数据集ISIA Food- 500，其中包含维基百科列表中的500个类别和399，726张图像，这是一个更全面的食品数据集，在类别覆盖率和数据量方面超过了现有的流行基准数据集。此外，我们提出了一个堆叠的全局-局部注意力网络，它由两个子网络的食物识别。一个子网络首先利用混合空间通道注意力来提取更多的区分特征，然后将这些多尺度区分特征从多个层聚合到全局级表示（例如，关于食物的质地和形状信息）。另一个生成注意区域（例如，成分相关区域），并进一步将这些来自不同层的多尺度区域特征聚合成地方级表示。这两种类型的功能，最终融合作为食品识别的综合表示。在ISIA Food-500和其他两个流行的基准数据集上的大量实验证明了我们所提出的方法的有效性，因此可以被认为是一个强大的基线。数据集、代码和模型可以在http://123.57.42.89/FoodComputing-Dataset/ISIA-Food500.html上找到。
英文摘要：
Food recognition has received more and more attention in the multimedia community for its various real-world applications, such as diet management and self-service restaurants. A large-scale ontology of food images is urgently needed for developing advanced large-scale food recognition algorithms, as well as for providing the benchmark dataset for such algorithms. To encourage further progress in food recognition, we introduce the dataset ISIA Food- 500 with 500 categories from the list in the Wikipedia and 399,726 images, a more comprehensive food dataset that surpasses existing popular benchmark datasets by category coverage and data volume. Furthermore, we propose a stacked global-local attention network, which consists of two sub-networks for food recognition. One subnetwork first utilizes hybrid spatial-channel attention to extract more discriminative features, and then aggregates these multi-scale discriminative features from multiple layers into global-level representation (e.g., texture and shape information about food). The other one generates attentional regions (e.g., ingredient relevant regions) from different regions via cascaded spatial transformers, and further aggregates these multi-scale regional features from different layers into local-level representation. These two types of features are finally fused as comprehensive representation for food recognition. Extensive experiments on ISIA Food-500 and other two popular benchmark datasets demonstrate the effectiveness of our proposed method, and thus can be considered as one strong baseline. The dataset, code and models can be found at http://123.57.42.89/FoodComputing-Dataset/ISIA-Food500.html.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
