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: Passenger action detection refers to the use of technical means to monitor passengers' abnormal or potentially dangerous behaviors, preventing accidents and ensuring the safety of passengers ...

Abstract: DETR-based methods have shown impressive performance in object detection tasks. The original DETR employs one-to-one sparse supervision, resulting in poor supervision capability. Denoising ...