A Lightweight High-Resolution Representation Backbone For Real-Time Keypoint-Based Object Detection

Detection of Road Objects With Small Appearance in Images for Autonomous Driving in Various Traffic Situations Using a Deep Learning Based Approach

“…This method can output object bounding box directly from the network without additional post-processing (e.g., matching key points, non-maximum suppression). As frequently reported in previous studies [10,[26][27][28], the key point strategy in CenterNet is more reasonable and interpretable with better detection performance and faster computational speed on object detection tasks than other approaches such as CornerNet and ExtremeNet. Therefore, we only compared the results of our proposed method with the basic CenterNet and one of the most popular anchor-based methods (i.e., YOLOv3) to show the superiority of our proposed method in this study.…”

DAN-NucNet++: A Bottom-Up Framework Using Dual Attention Network for Simultaneous Nuclei Segmentation and Classification in Digital Histology Images

“…Keypoint-based methods produce an enormous amount of incorrect bounding boxes [17]. Furthermore, keypoint-based methods also have reduced inference speed [19].…”