A US fighter jet pilot rescued by special forces after being shot down over Iran in April described a shocking sight before ejecting from his aircraft: multiple Iranian drones hovering in the air, ...

Background & Motivation Background: Open-Vocabulary Object Detection (OVOD) requires detectors to recognize categories not annotated during training. The mainstream approach leverages the cross-modal ...

Abstract: Underwater salient object detection (USOD) faces two major challenges that hinder accurate detection: substantial image noise owing to water turbidity and low foreground-background contrast ...

To say 2025 was a busy news year would be an understatement. There were wildfires and hurricanes and plenty of political intrigue. We also enjoyed some memorable moments in sports and entertainment.

This study proposes a feature-enhanced Faster R-CNN algorithm to address the challenges of dense targets, varying object shapes, and complex backgrounds in remote sensing image detection. First, the ...

Training-free framework that converts SAM3 into a real-time multi-class open-vocabulary detector. Achieves 55.8 AP on COCO val2017 (80 classes) at 15.8 FPS (4 classes, 1008px) on a single RTX 4080.