Logo Net

by Steven C.H. Hoi,Xiongwei Wu,Hantang Liu,Yue Wu,Huiqiong Wang,Hui Xue,Qiang WuUnknown

Logo Net

Logo detection is a special type of object detection in computer vision, which can be useful for many applications. Here we introduce a new logo database for brand recognition called LOGO-Net, with a total of 100 brands, 160 logo categories, 73414 images, and 130,608 logo objects annotated with a bounding box and category label. Using Deep Region-based convolutional neural network (DRCN), we learn logo object detectors and apply them for brand recognition tasks, and establish new state of the art on logo detection benchmarks on real-world product images. Here we provide the LOGO-Net database and the trained collections of R-CNNs for academic research and education purposes.

Dataset Attributes

Label SVG
TasksDetection
Label SVG
CategoriesLogos
Label SVG
SensorWeb Sampling