Abstract: Object detection using visible-infrared images has become increasingly crucial for all-day applications of uncrewed aerial vehicles (UAVs). However, existing multimodal detection methods ...

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.

Abstract: Maritime environments often face visibility challenges due to haze which significantly impacts detection models. However, existing maritime object detection algorithms often neglect haze ...

This project combines a high-speed YOLOE vision engine running on a laptop (50+ FPS) with an ESP32 that wirelessly controls servos, sensors, and peripherals in real time. The result is a powerful dual ...