RF-DETR 是 Roboflow 推出的实时目标检测与实例分割模型架构,在 COCO 上达到 SOTA,适合直接微调落地。
基于 PyTorch 的目标检测工具箱与基准库,集成了 Faster R-CNN、DETR、Grounding DINO 等主流模型,方便训练、评测和复现检测与分割算法。
一套可直接在 Colab 跑的计算机视觉教程合集,覆盖 ResNet 到 RF-DETR、YOLO11、SAM 3、Qwen3-VL 等模型,帮你快速上手训练、微调与推理。
基于飞桨的目标检测工具库,覆盖检测、实例分割、多目标跟踪和实时多人关键点检测,提供从训练到部署的完整流程。
用 TensorRT 网络定义 API 手写实现 YOLO、ResNet、Swin Transformer 等主流深度学习模型,并附 C++ 推理代码,方便直接部署到 NVIDIA GPU。
HuggingFace Transformers 库的实战教程合集,用 Jupyter Notebook 演示 BERT、GPT-2、LayoutLM、ViT 等模型的微调与推理,适合边跑边学。
RF-DETR 1.10.1 is a patch release: compile=True now actually reaches CUDA's torch.compile path under the default multi scale=True recipe, and peak CUDA memory in segmenta
…https://github.com/jamjamjon/usls/pull/160 Add RT-DETRv4 by @jamjamjon in https://github.com/jamjamjon/usls/pull/161 New Contributors @numfin made their first contributio
…ithms and Benchmarks, including: - Support [Focus-DETR (ICCV'2023)](https://github.com/IDEA-Research/detrex/tree/main/projects/focus detr) - Support [SQR-DETR (CVPR'2023)
…eXt-1K pretrained backbones. - Release Deformable-DETR-R50 pretrained weights. - Release DETA and better H-DETR pretrained weights: achieving 50.2 AP and 49.1 AP respecti
What's New New Algorithms - Support Anchor-DETR - Support DETA More training techniques - Support EMAHook during training by setting train.model ema.enabled=True , which
…provides us a detailed usage about onnx export in detrex. Please see this issue https://github.com/IDEA-Research/detrex/issues/192 What's New New Algorithm - MaskDINO COC