Portal de Periódicos da CAPES

Real-Time Logo Detection in Brand-Related Social Media Images

2019; Springer Science+Business Media; Linguagem: Inglês

10.1007/978-3-030-20518-8_11

1611-3349

Oscar Orti, Rubén Tous, Mauro Gomez, Jonatan Poveda, Leonel Cruz, Otto Wust,

Visual Attention and Saliency Detection

This paper presents a work consisting in using deep convolutional neural networks (CNNs) for real-time logo detection in brand-related social media images. The final goal is to facilitate searching and discovering user-generated content (UGC) with potential value for digital marketing tasks. The images are captured in real time and automatically annotated with two CNNs designed for object detection, SSD InceptionV2 and Faster Atrous InceptionV4 (that provides better performance on small objects). We report experiments with 2 real brands, Estrella Damm and Futbol Club Barcelona. We examine the impact of different configurations and derive conclusions aiming to pave the way towards systematic and optimized methodologies for automatic logo detection in UGC.