These systems depend heavily on data annotation to train AI/ML models that enable real-time perception and decision-making. Raw data is collected from a suite of sensors, including LiDAR, radar, ...

Self-training is a novel learning paradigm that generates pseudolabels for unlabeled data, enabling deep learning models to be trained without the need for humanlabeled data. This article proposes ...

To address the challenge of effectively classifying and recycling construction and demolition waste (CDW), this study proposes a lightweight object detection algorithm, CDWNet, based on YOLO11. The ...