RF-DETR is a real-time, transformer-based object detection model architecture developed by Roboflow and released under the Apache 2.0 license. RF-DETR is the first real-time model to exceed 60 AP on ...
Abstract: Object detection is a cornerstone of environmental perception in advanced driver assistance systems (ADASs). However, most existing methods rely on RGB cameras, which suffer significant ...
Abstract: Real-time object detection on edge devices operates under tight computational, memory, and power budgets. Prior work typically treats model compression and hardware acceleration ...
The dataset is already organized in YOLO format in the steel_dataset/ directory. If you need to reorganize from original format, see utility/reorganize_dataset.py. steel-defect-detection/ ├── ...