DeepLogo: Hitting Logo Recognition with the Deep Neural Network Hammer
This work addresses logo recognition for industrial applications like brand tracking, but it is incremental as it applies existing DCNN methods to a specific domain.
The paper tackled logo recognition by applying deep convolutional neural networks (DCNNs), achieving state-of-the-art accuracy on a popular dataset.
Recently, there has been a flurry of industrial activity around logo recognition, such as Ditto's service for marketers to track their brands in user-generated images, and LogoGrab's mobile app platform for logo recognition. However, relatively little academic or open-source logo recognition progress has been made in the last four years. Meanwhile, deep convolutional neural networks (DCNNs) have revolutionized a broad range of object recognition applications. In this work, we apply DCNNs to logo recognition. We propose several DCNN architectures, with which we surpass published state-of-art accuracy on a popular logo recognition dataset.