Abstract: Monocular 3D object detection is challenging due to the lack of accurate depth. However, existing depth-assisted solutions still exhibit inferior performance, whose reason is universally ...

Abstract: In LiDAR-based 3D object detection, severe occlusion often leads to incomplete object shape in point cloud scene. It is a critical challenge as it will significantly degrade detection ...

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.

HOI-DETR is a transformer-based framework for detecting hands, hand-held objects, and their interactions in images and video. Built on the Co-DETR architecture, it adds a lightweight interaction ...