Logo detection and brand recognition with one-stage logo detection framework and simplified resnet50 backbone

Abstract

Logo and brand name are two concepts which are typically studied in many course subjects. In education context, automated logo detection and brand name recognition from digital image or video are very crucial as a learning tool to achieve learning outcomes. One technical issue in the logo detection and brand name recognition is its requirement to develop model that achive fast recognition speed and high recognition accuracy. One-stage detector is a breakthrough and innovative object detection framework; however, long duration and computing power required to carry out training and detection processes using the backbone deep architecture are often considered to be the challenge of this framework. The objective of this study is to propose a novel ResNet variant models using ResNet-50 as the basis. The empiric results showed that the model 2 achieved 0.408 mAP, the best average accuracy with training time 1.41 hour. The original Resnet50 model, in contrast, achieved 0.556 mAP average accuracy with 1.91 hour training time. The detection testing of the proposed Model 2 model was 23.47 fps, while the detection testing of Resnet50 model was 29.33 Fps.