ChineseFoodNet: A large-scale Image Dataset for Chinese Food Recognition

Abstract

In this paper, we introduce a new and challenging large-scale food image dataset called "ChineseFoodNet", which aims to automatically recognizing pictured Chinese dishes. Most of the existing food image datasets collected food images either from recipe pictures or selfie. In our dataset, images of each food category of our dataset consists of not only web recipe and menu pictures but photos taken from real dishes, recipe and menu as well. ChineseFoodNet contains over 180,000 food photos of 208 categories, with each category covering a large variations in presentations of same Chinese food. We present our efforts to build this large-scale image dataset, including food category selection, data collection, and data clean and label, in particular how to use machine learning methods to reduce manual labeling work that is an expensive process. We share a detailed benchmark of several state-of-the-art deep convolutional neural networks (CNNs) on ChineseFoodNet. We further propose a novel two-step data fusion approach referred as "TastyNet", which combines prediction results from different CNNs with voting method. Our proposed approach achieves top-1 accuracies of 81.43% on the validation set and 81.55% on the test set, respectively. The latest dataset is public available for research and can be achieved at https://sites.google.com/view/chinesefoodnet.

8 pages, 5 figure, 2 tables

References in corpus (4)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Deep Learning in Neural Networks: An Overview
• SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size
• DenseNet: Implementing Efficient ConvNet Descriptor Pyramids

Cited by in corpus (10)

• FoodX-251: A Dataset for Fine-grained Food Classification
• Leveraging Automatic Personalised Nutrition: Food Image Recognition Benchmark and Dataset based on Nutrition Taxonomy
• Deep Cooking: Predicting Relative Food Ingredient Amounts from Images
• ISIA Food-500: A Dataset for Large-Scale Food Recognition via Stacked Global-Local Attention Network
• Dividing and Conquering Cross-Modal Recipe Retrieval: from Nearest Neighbours Baselines to SoTA
• Picture-to-Amount (PITA): Predicting Relative Ingredient Amounts from Food Images
• Spatial-Aware Non-Local Attention for Fashion Landmark Detection
• Deep Learning based Food Instance Segmentation using Synthetic Data
• FooDI-ML: a large multi-language dataset of food, drinks and groceries images and descriptions
• Revamping Cross-Modal Recipe Retrieval with Hierarchical Transformers and Self-supervised Learning