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.

We propose MaskCut approach to generate pseudo-masks for multiple objects in an image. CutLER can learn unsupervised object detectors and instance segmentors solely on ImageNet-1K. CutLER exhibits ...

A new self-destructing backdoor called Mistic used in intrusions since April appears to be linked to a criminal gang that ...

点击上方“Deephub Imba”,关注公众号,好文章不错过 !微调LocateAnything-3B，实现当图像中有 300+ 个密集重叠目标、人工标注不可行时的实用方案。假设手头有一批种子发芽托盘、谷物质检图像或植物学调查照片。每张图像包含 100–500+ ...

A Florida sheriff's office has an unlikely officer helping them solve cases and hunt down murder weapons. Meet Splash, a 2½-year-old male otter, trained to assist authorities in search-and-recovery ...