Automatic Dataset Augmentation

by   Yalong Bai, et al.
Harbin Institute of Technology
ByteDance Inc.

Large scale image dataset and deep convolutional neural network (DCNN) are two primary driving forces for the rapid progress made in generic object recognition tasks in recent years. While lots of network architectures have been continuously designed to pursue lower error rates, few efforts are devoted to enlarge existing datasets due to high labeling cost and unfair comparison issues. In this paper, we aim to achieve lower error rate by augmenting existing datasets in an automatic manner. Our method leverages both Web and DCNN, where Web provides massive images with rich contextual information, and DCNN replaces human to automatically label images under guidance of Web contextual information. Experiments show our method can automatically scale up existing datasets significantly from billions web pages with high accuracy, and significantly improve the performance on object recognition tasks by using the automatically augmented datasets, which demonstrates that more supervisory information has been automatically gathered from the Web. Both the dataset and models trained on the dataset are made publicly available.


page 4

page 5

page 6

page 10


Learning Scene Gist with Convolutional Neural Networks to Improve Object Recognition

Advancements in convolutional neural networks (CNNs) have made significa...

Using Web Co-occurrence Statistics for Improving Image Categorization

Object recognition and localization are important tasks in computer visi...

Self-informed neural network structure learning

We study the problem of large scale, multi-label visual recognition with...

Webpage Segmentation for Extracting Images and Their Surrounding Contextual Information

Web images come in hand with valuable contextual information. Although t...

SdcNet: A Computation-Efficient CNN for Object Recognition

Extracting features from a huge amount of data for object recognition is...

Image understanding and the web

The contextual information of Web images is investigated to address the ...

Context Augmentation for Convolutional Neural Networks

Recent enhancements of deep convolutional neural networks (ConvNets) emp...

Please sign up or login with your details

Forgot password? Click here to reset