Traffic Sign Recognition Using Ulam's Game
Haofeng Zheng, Ruikang Luo, Yaofeng Song, Yao Zhou, Jiawen Yu · 2022 17th International Conference on Control, Automation, Robotics and Vision (ICARCV) · 2022
In this paper, we propose a conditional early exiting framework with Ulam's Game for traffic sign recognition. Since the traffic sign recognition system has extremely high requirements on dynamic performance, we pays more attention to improving the detection efficiency, hoping to obtain results in a shorter time. In our system, we use a modified ResNet-50 as backbone network to do feature extraction and use a Pooling module to accumulate feature. Then, we have a Gate module to determine whether the feature have accumulated enough to begin Ulam's Game. A classifier is used to get candidate results, which are used to run Ulam's Game and get the final prediction. The model shows good detection accuracy and dynamic performance in multiple data sets (Mini-Kinetics, ActivityNet, Lisa).