New AI Model Spots Tiny Drones Hiding in Cluttered Skies
Researchers have developed FUS-DETR, a transformer-based AI detector that achieves state-of-the-art accuracy in spotting small drones against cluttered backgrounds in ...
Researchers have developed FUS-DETR, a transformer-based AI detector that achieves state-of-the-art accuracy in spotting small drones against cluttered backgrounds in ...
A new prediction-based neural network called RMTA-Net combines recurrent temporal processing, learnable memory banks, and adaptive attention to detect video ...
Researchers compared seven convolutional neural network architectures for detecting diseases and pests in star fruit and found that the classic ...
Researchers have developed LMANet, a lightweight multiscale attention network that achieves high-accuracy skin-lesion segmentation with only 5.6 million parameters by ...
Researchers in Chongqing have developed a wavelet-guided deep learning framework that coordinates frequency-domain and spatial cues to sharply localize changes ...
Researchers at Hohai University have developed ShuffleNet-MSAA, a lightweight AI system that uses adaptive attention and sonar imaging to detect ...
Researchers have built a drone perception framework that pairs an enhanced YOLOv7 detector with graph attention network reasoning and spatial ...
Researchers have developed AttentiveGaze, a real-time multimodal AI framework that fuses face, eye, and head pose cues with attention mechanisms ...
Researchers have developed a real-time computer vision system that reads students' facial expressions during online lessons and converts four learning-centered ...
A new self-supervised framework called AnomalyMHKD combines self-distillation, cross-architecture distillation and dynamic teacher freezing to detect industrial anomalies with up ...
© 2026 Scienmag - Science Magazine
© 2026 Scienmag - Science Magazine